VISVESVARAYA TECHNOLOGICAL UNIVERSITY

Belagavi-590 018, Karnataka

“THE INTERNSHIP REPORT”

ON
“Industry Internship”

Submitted in Partial Fulfillment for the Award of Degree of

BACHELOR OF ENGINEERING
in
COMPUTER SCIENCE AND ENGINEERING
Submitted by
SANJANA RAWAL D
1BH22CS096

Under the Guidance of

Internal Guide: External Guide:
Mr. Ratikanta Majhi Mr. Karthiksaran D
Associate Professor Senior AI Engineer
Dept. of Computer Science and Engineering iSpark Learning Solutions Pvt. Ltd.

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

BANGALORE TECHNOLOGICAL INSTITUTE

(NAAC Accredited & An ISO 9001:2015 Certified Institute)
Kodathi Village, Varthoor Hobli, Bangalore East Tq, Bangalore Urban District,
Bangalore, 560 035, Karnataka

2025-26
 Department of Computer Science and Engineering

CERTIFICATE

Certified that the Industry Internship work entitled “AI Engineer Accelerator Program: Build
Real-World AI Projects in 4 Industries (REMOTE)” carried out by Sanjana Rawal D bearing
USN: 1BH22CS096 Bonafide student of Bangalore Technological Institute, Bengaluru in
partial fulfilment for the award of Bachelor of Engineering in Computer Science and
Engineering of the Visvesvaraya Technological University, Belagavi during the academic year
2025-26. It is certified that all corrections/suggestions indicated for the Internal Assessment have
been incorporated in the Internship Report. The internship report has been approved as it satisfies
the academic requirements in respect of internship prescribed for the said degree.

Dr. Payal Koolwal Dr. D. Jerline Sheebha Anni Dr. H S Nanda

Professor Professor & Head
Principal
Internship Coordinator Dept. of CSE

EXTERNAL EXAMINATION

Name of the Examiners Signature with Exam Date

1. 1.

2. 2.
 Department of Computer Science and Engineering

CERTIFICATE
"Certified that this Internship report is an original report of work done by me under the guidance

of Internship Mentor Mr. Ratikanta Majhi and under the supervision of Internship

Supervisor(lndustry) Mr. Karthiksaran D submitted as a part of the Internship Course of

Undergraduate Programme of Visvesvaraya Technological University, Jnana Sangama, Machhe,

Belagavi-590018".

Date:
Signature of the student

Sign of Sign. of
Department internship Co-ordinator College Internship Co-ordinator
Dr. Payal Koolwal Dr. Mohamed Saleem
COMPANY CERTIFICATE
 ACKNOWLEDGEMENT
The successful completion of any task would be incomplete without mentioning the people who
made it possible with constant guidance and encouragement leading to success.

Guidance and deadlines play a very important role in the successful completion of the
internship on time, and I would like to express my sincere gratitude to my Internship Guide,
Mr. Ratikanta Majhi, Associate Professor, Department of Computer Science and
Engineering, and my mentor, Mr. Karthiksaran D, Internship Coordinator, iSpark Learning
Solutions Pvt. Ltd., for their unfailing encouragement and valuable suggestions.

A special thanks to the Management of iSpark Learning Solutions, for providing me an

internship in your esteemed organization.

Guidance and deadlines play a very important role in successful completion of the Internship
on time. The gratitude conveyed to my Internship Co-Ordinator Dr. Payal Koolwal,
Professor, Department of Computer Science and Engineering, for her unfailing encouragement
and suggestions.

I express my gratitude to Dr. D. Jerline Sheebha Anni, HOD, Department of Computer

Science and Engineering, for her unfailing encouragement and suggestions in the course of
Internship work.

I take this opportunity to express my gratitude to our Principal Dr. H S Nanda for his
constant support and encouragement.

I am grateful to our President Dr. A Prabhakara Reddy for having provided us with a
constant support and encouragement.

Finally, a note of thanks to the Department of Computer Science and Engineering, both
teaching and non-teaching staffs for their cooperation extended to me.

Last but not the least, I Also acknowledge the support and the feedback of my parents and
friends of all time.

Sanjana Rawal D
1BH22CS096

i
 ABSTRACT

This internship at iSpark Learning Solutions provided a comprehensive, industry-oriented, and

practically focused learning experience through the AI Engineer Accelerator Program: Build
Real-World AI Projects in 4 Industries (REMOTE), emphasizing modern Artificial
Intelligence technologies, AI engineering workflows, implementation-based learning, and real-
world problem-solving methodologies. The internship combined structured technical modules,
guided assignments, quizzes, hands-on implementation activities, project-based learning, and
practical evaluations to strengthen both theoretical understanding and technical proficiency in
key domains such as Machine Learning, Deep Learning, recommendation systems, Generative
AI, backend automation, AI deployment workflows, and production-ready AI systems. A major
component of the internship was the design, development, and study of a capstone project titled
“Retail AI Decision Assistant,” an AI-powered retail analytics and recommendation platform
developed to enhance customer personalization, improve recommendation accuracy, optimize
demand forecasting, and support intelligent business decision-making processes. The system
incorporated recommendation engines, forecasting models, analytics dashboards, visualization
components, and AI-driven decision-support functionalities using modern AI engineering
concepts, data processing pipelines, and visualization tools. Through this project, practical
exposure was gained in areas including recommendation architectures, machine learning
workflows, evaluation metrics, backend integration concepts, data preprocessing techniques, and
intelligent analytics systems commonly used in real-world retail and business environments. The
internship also introduced modern development tools, deployment methodologies, and cloud-
oriented AI engineering practices through implementation-focused sessions and practical
activities. Technologies and platforms such as TensorFlow, Keras, Streamlit, Docker, and
FastAPI concepts were explored extensively during the program to understand AI model
development, deployment, and scalable application workflows. In addition to technical learning,
the internship also focused on professional development through sessions on resume preparation,
portfolio building, LinkedIn optimization, communication skills, presentation techniques, and
interview readiness, contributing significantly toward career preparation and industry awareness.
Overall, the internship enhanced technical competency, analytical thinking, implementation
capability, teamwork, problem-solving skills, and professional confidence, thereby establishing a
strong foundation for future academic research, software development projects, and career
opportunities in the field of Artificial Intelligence and AI engineering.

ii
 INDEX
TITLE PAGE NO.
Acknowledgment i
Abstract ii
Contents iii

CONTENTS
SI. No. Chapter Pages No.

1. About the Organization 1

1.1 Introduction 1
1.2 Vision of the Organization 2
1.3 Core Values of the Organization 2
1.4 Organizational Background and Growth 3
1.5 Services and Offerings 3
1.6 Impact and Achievements 4
1.7 Organizational Structure and Work Environment 5
2. Objectives of Internship 6
2.1 Primary Objectives of the Internship 6
2.2 Organizational Objectives 7
2.3 Technical Objectives 8
2.4 Professional and Career Objectives 9
2.5 Project-Specific Objectives 9
3. Learning Experiences 11
3.1 Technical Learning 11
3.2 Tools and Technologies 12
3.3 Project-Based Learning Experience 13
3.4 Soft Skills and Professional Development 14
3.5 Problem-Solving and Analytical Thinking 15
3.6 Practical Implementation 16
4. Learning Outcomes 17
4.1 Practical Knowledge 17
4.2 Industry Exposure and Work Culture Understanding 19

iii
 4.3 Achievements and Certifications 20
4.4 AI Engineering and Sustem Design Learning 21
4.5 Main Project Outcome: Retail AI Decision Assistant 23
4.6 Professional Development and Career Readiness 25
4.7 Overall Internship Outcome 26
5. CONCLUSION 27
APPENDIX A: CODE 29
APPENDIX B: SCREENSHOT 45

iv
Industry Internship 2025-26

Chapter 1
ABOUT THE ORAGANIZATION

1.1 Introduction

iSpark Learning Solutions is an education-focused technology organization dedicated to

transforming conventional learning into a practical, skill-oriented, and industry-relevant
experience. The organization primarily focuses on empowering students with modern
technological competencies through structured training programs, experiential learning
methodologies, and exposure to emerging technologies such as Artificial Intelligence (AI),
Machine Learning (ML), Robotics, Internet of Things (IoT), Coding, and Cyber Security.

In the rapidly evolving digital era, industries increasingly demand professionals who possess both
strong theoretical understanding and practical implementation skills. Traditional educational
systems often emphasize conceptual learning while providing limited exposure to real-world
applications, resulting in a gap between academic education and industry expectations. iSpark
Learning Solutions was established with the objective of bridging this gap by delivering
industry-aligned technical education and hands-on learning opportunities.

The organization follows a structured and application-oriented approach to learning, where

students actively participate in technical modules, assignments, quizzes, projects, and practical
exercises. Through this methodology, learners are encouraged to apply theoretical concepts to
solve practical problems, thereby improving technical proficiency, analytical thinking, and
problem-solving abilities.

A key strength of the organization is its focus on emerging technologies and future-ready skills.
Students are provided exposure to Artificial Intelligence, Deep Learning, Natural Language
Processing (NLP), Computer Vision, Generative AI, cloud-based systems, and modern AI
engineering workflows. The organization also integrates practical development tools such as
TensorFlow, Keras, and advanced AI frameworks to ensure alignment with current industry
standards.

Apart from technical learning, iSpark Learning Solutions emphasizes professional development
and discipline through continuous assessments, structured learning schedules, self-paced modules,
and industry-oriented project work. This holistic approach enables students to enhance both
technical competence and professional readiness, preparing them effectively.

Department of AI&ML, BTI, Bengaluru-35 1

Industry Internship 2025-26
1.2 Vision of the Organization

The vision of iSpark learning Solutions is to become a leading platform for skill-based technical
education by bridging the gap between academic learning and industry requirements through
innovative and practical learning methodologies.

The organization aims to create a future-ready generation of students equipped with advanced
technical skills, industry exposure, and confidence to adapt to rapidly changing technological
environments. By integrating modern technologies such as Artificial Intelligence, Machine
Learning, Robotics, and Generative AI into its educational ecosystem, the organization seeks to
redefine traditional learning and make education more practical, engaging, and outcome-
oriented.

A significant aspect of this vision is the promotion of experiential and hands-on learning,
enabling students to apply theoretical concepts to real-world applications. The organization also
strives to create a collaborative ecosystem connecting students, educational institutions, mentors,
and industry professionals, thereby facilitating continuous learning, innovation, and professional
growth.

Ultimately, the vision of iSpark Learning Solutions is to empower learners with the technical
expertise, creativity, and problem-solving capabilities required to succeed in a competitive global
environment.

1.3 Core Values of the Organization

The core values of iSpark Learning Solutions form the foundation of its educational philosophy
and organizational culture. These values guide the organization in delivering quality education
and fostering professional development among learners.
• Excellence: The organization is committed to providing high-quality technical education
through structured learning programs, practical exposure, and continuous improvement in
training methodologies.
• Innovation: iSpark Learning Solutions encourages innovation by integrating emerging
technologies such as Artificial Intelligence, Machine Learning, Robotics, and Generative
AI into its programs, promoting creative problem-solving and technological
advancement.
• Practical Learning: The organization strongly believes in experiential learning, where
students gain hands-on experience through assignments, projects, and real-world
applications rather than relying solely on theoretical concepts.
• Collaboration: A collaborative learning environment is promoted through interaction

Department of AI&ML, BTI, Bengaluru-35 2

Industry Internship 2025-26
between students, mentors, institutions, and industry experts, enabling knowledge sharing
and teamwork.
• Impact: The organization focuses on enhancing employability, technical competence,
and professional readiness, ensuring that learners are capable of meeting industry
expectations effectively.
These values collectively ensure that the organization maintains a balanced approach toward
academic learning, practical implementation, innovation, and professional development.

1.4 Organizational Background and Growth

iSpark Learning Solutions was established with the objective of modernizing technical education
and preparing students for future technologies through skill-based learning programs. Since its
inception, the organization has focused on integrating practical exposure with academic
knowledge to develop industry-ready professionals.
The organization has consistently expanded its educational initiatives by introducing STEM-based
learning programs, Artificial Intelligence training, robotics education, coding workshops, and AI-
powered educational tools. It has collaborated with schools, colleges, and universities to promote 21st-
century technical skills among students.
Key Milestones:
• Establishment as a technology-driven EdTech organization focused on STEM education
• Introduction of AI-powered educational systems and humanoid AI teacher robots
• Expansion of training programs in Artificial Intelligence, Robotics, IoT, and Coding
• Collaboration with schools, colleges, and universities across multiple regions
• Successful training of over 16,000 students
• Partnership with more than 41 educational institutions
• Launch of industry-oriented AI Engineering Accelerator Programs
• Integration of practical AI projects and real-world industry applications into learning
modules
This growth reflects the organization’s commitment to innovation, scalability, and practical
technical education.

1.5 Services and Offerings

iSpark Learning Solutions provides a wide range of educational and technical services designed
to improve students’ practical knowledge, technical expertise, and industry readiness. The
organization combines theoretical instruction with hands-on implementation to create a
comprehensive learning ecosystem.
Key Services and Offerings:
• Artificial Intelligence and Machine Learning Programs: Training programs focused
Department of AI&ML, BTI, Bengaluru-35 3
Industry Internship 2025-26
on AI, Machine Learning, Deep Learning, NLP, Computer Vision, and Generative AI
technologies.
• STEM Education Programs: Courses and workshops in Robotics, Internet of Things
(IoT), Coding, Digital Literacy, Cyber Security, and 3D Printing.
• Industry-Oriented Internship Programs: Structured internships that provide practical
exposure through assignments, projects, quizzes, and real-world AI applications.
• Hands-on AI Projects: Opportunities to develop AI-based systems and solutions using
modern frameworks such as TensorFlow and Keras.
• AI-Powered Educational Tools: Development and implementation of AI-based learning
systems and humanoid AI teacher robots to enhance interactive learning experiences.
• Skill Development Workshops: Sessions aimed at improving communication skills,
professional discipline, problem-solving abilities, and career readiness.
• Industry Collaboration and Institutional Partnerships: Collaboration with
educational institutions to promote skill-based learning and industry integration.
• Research and Innovation Support: Encouragement of experimentation and innovation
through practical projects and modern technology platforms.
These services collectively help students develop technical competence, practical exposure, and
professional confidence required for modern technology careers.

1.6 Impact and Achievements

iSpark Learning Solutions has made a significant contribution to technical education and skill
development by promoting practical learning and industry-oriented training methodologies. The
organization has positively impacted students by equipping them with modern technical skills
and real-world problem-solving capabilities.
Key Achievements:
• Training of more than 10,000 students
• Training of more than 16,000 students
• Collaboration with over 41 schools, colleges, and universities
• Development of AI-powered educational tools and humanoid AI systems
• Successful implementation of STEM and AI learning programs
• Introduction of industry-focused AI Engineering Accelerator Programs
• Practical exposure to emerging technologies such as AI, ML, NLP, Computer Vision, and
Generative AI
Overall Impact:
• Enhanced student employability through practical technical training
• Improved understanding of Artificial Intelligence and modern technologies
• Promotion of innovation, analytical thinking, and problem-solving skills
Department of AI&ML, BTI, Bengaluru-35 4
Industry Internship 2025-26
• Development of industry-ready technical competencies among learners
• Bridging the gap between academic learning and professional industry requirements

1.7 Organizational Structure and Work Environment

iSpark Learning Solutions follows a structured and learner-centric organizational framework that
promotes innovation, collaboration, and practical skill development. The organizational structure
includes leadership teams, technical mentors, program coordinators, and support staff who
collectively ensure smooth execution of educational programs and internship activities.
The work environment is designed to simulate real-world professional learning conditions while
maintaining flexibility and accessibility for students through online learning platforms and structured
modules. During the internship program, students actively participated in daily learning activities,
quizzes, assignments, and technical exercises, encouraging continuous engagement and disciplined
learning practices.
Key Features of Work Environment:
• Structured Learning Approach: Well-organized modules, assignments, quizzes, and
phased learning schedules ensured systematic skill development.
• Practical and Industry-Oriented Training: Focus on real-world AI concepts, AI
engineering workflows, and modern technology applications.
• Mentorship and Guidance: Continuous support from mentors and technical instructors
throughout the internship program.
• Collaborative and Technology-Driven Environment: Use of modern AI tools, cloud-
based learning systems, and practical development platforms such as TensorFlow and
Keras.
• Self-Paced and Remote Learning: Flexibility to learn remotely while maintaining
consistency through regular assessments and deadlines.
The overall work environment encouraged continuous learning, technical exploration, and
professional development, helping students gain confidence and industry-oriented experience
throughout the internship program.

Department of AI&ML, BTI, Bengaluru-35 5

Industry Internship 2025-26
Chapter 2
OBJECTIVES OF INTERNSHIP

The internship at iSpark learning Solutions was designed as a structured and industry-oriented
training program aimed at strengthening practical knowledge in Artificial Intelligence, Machine
Learning, Deep Learning, and modern AI engineering concepts. In today’s rapidly evolving
technological landscape, it is essential for students to develop not only theoretical understanding
but also practical exposure to industry-relevant tools, workflows, and problem-solving
methodologies. The internship provided learners with systematic training in AI concepts, real-
world AI applications, and modern development practices, enabling them to bridge the gap
between academic learning and industry expectations.

The program followed a phased and experiential learning methodology, where students
progressed through multiple modules covering AI foundations, machine learning concepts,
neural networks, Natural Language Processing (NLP), Computer Vision, Generative AI, MLOps,
and production AI systems. Through daily lessons, quizzes, assignments, and practical activities,
students gained hands-on exposure to the workflow involved in designing, evaluating, and
implementing AI-based systems.

The internship also provided exposure to modern AI frameworks and development tools such as
TensorFlow and Keras, enabling learners to understand how AI models are developed, trained,
and evaluated in practical environments. In addition, the program emphasized discipline,
consistency, self-learning, and professional development through continuous assessments and
structured learning schedules. Overall, the internship objectives focused on holistic development
by combining technical expertise, practical implementation, analytical thinking, and professional
readiness.

2.1 Primary Objectives of the Internship

The primary objectives of the internship were focused on developing a strong foundation in
Artificial Intelligence and AI engineering practices through practical and application-oriented
learning. The internship at iSpark learning Solutions was structured to provide students with
exposure to modern AI technologies, workflows, and real-world industry concepts, ensuring
alignment with current technological trends and professional requirements.

The internship emphasized experiential learning through structured modules, quizzes,

assignments, and practical exercises. This approach ensured that students actively engaged with
concepts rather than limiting their learning to theoretical understanding. Through continuous

Department of AI&ML, BTI, Bengaluru-35 6

Industry Internship 2025-26
learning activities and exposu
Insight into how software solutions are developed, tested, and deployed in professional
environments. This helped in building confidence and preparing them to handle real-world
challenges effectively.
Key Primary Objectives:
• Building Foundations in Artificial Intelligence: Developing conceptual understanding
of Artificial Intelligence, Machine Learning, and Deep Learning along with their real-
world applications.
• Understanding AI Engineering Workflows: Learning the stages involved in AI system
development, including data collection, model training, evaluation, optimization, and
deployment.
• Practical Exposure to AI Technologies: Gaining hands-on knowledge of AI frameworks
and tools such as TensorFlow and Keras.
• Developing Problem-Solving Skills: Enhancing analytical thinking and the ability to
approach technical challenges systematically.
• Understanding Modern AI Applications: Exploring areas such as Natural Language
Processing, Computer Vision, Generative AI, and AI-powered systems.
• Exposure to Real-World AI Projects: Understanding how AI concepts are applied across
multiple industries including healthcare, finance, retail, and manufacturing.
• Improving Technical Discipline: Developing consistency, self-learning habits, and time
management through structured daily learning activities and assessments.

2.2 Organizational Objectives

The organizational objectives of the internship reflect the broader mission of iSpark learning
Solutions in promoting practical, industry-oriented, and future-ready technical education. These
objectives focus on preparing students not only academically but also professionally by
equipping them with relevant technical skills and exposure to emerging technologies.

The organization aims to bridge the gap between conventional education and industry
requirements by integrating practical learning methodologies, AI-focused training, and real-
world applications into its programs. Through structured learning phases and continuous
assessments, the organization encourages innovation, technical growth, and independent learning
among students.

Additionally, the organizational objectives emphasize employability enhancement, technical

competence, and preparation for future technology-driven careers. By introducing learners to
modern AI systems and industry practices, the organization ensures that students remain aligned
with evolving technological advancements.
Department of AI&ML, BTI, Bengaluru-35 7
Industry Internship 2025-26
Key Organizational Objectives:
• Bridging Academia and Industry: Reducing the gap between theoretical concepts and
practical implementation through experiential learning.
• Developing Industry-Ready AI Professionals: Equipping students with AI engineering
knowledge, technical skills, and practical exposure relevant to professional environments.
• Promoting Practical and Project-Based Learning: Encouraging active participation in
assignments, quizzes, and AI-related practical activities.
• Integrating Emerging Technologies: Providing exposure to Artificial Intelligence,
Deep Learning, Generative AI, NLP, and modern AI engineering concepts.
• Enhancing Employability: Improving technical competence and professional readiness
through structured learning programs.
• Encouraging Innovation and Research: Motivating students to explore advanced AI
technologies and innovative problem-solving approaches.
• Supporting Career Development: Guiding students toward future opportunities in AI
engineering and technology-driven industries.
• Building a Collaborative Ecosystem: Creating interaction between learners, mentors,
institutions, and technology professionals for effective knowledge sharing.

2.3 Technical Objectives

The technical objectives of the internship were aimed at developing practical knowledge in
Artificial Intelligence, Machine Learning, Deep Learning, and AI engineering systems. The
internship provided learners with an environment where theoretical concepts could be connected
with practical implementation, thereby improving technical understanding and real-world
problem-solving capabilities.

Students were exposed to various aspects of application development, including front-end design,
backend integration, database management, and API usage. This comprehensive exposure ensured
a holistic understanding of how software systems function in real-world scenarios.
Additionally, the integration of advanced technologies such as Generative AI enabled students to
explore innovative approaches and develop intelligent applications, enhancing their technical
capabilities.
Key Technical Objectives:
• Artificial Intelligence Fundamentals: Developing mobile applications using Kotlin and
understanding lifecycle and architecture.
• User Interface (UI) Design: Creating responsive, user-friendly, and visually appealing
interfaces.

Department of AI&ML, BTI, Bengaluru-35 8

Industry Internship 2025-26
• Generative AI Integration: Learning concepts related to LLMs, Retrieval-Augmented
Generation (RAG), AI agents, and multimodal systems.
• Understanding Backend Concepts: Learning data flow, database connectivity, and
backend logic.
• Working with Development Tools: Gaining proficiency in Android Studio and cloud-
based platforms.
• Debugging and Testing: Identifying and resolving errors to ensure application stability.
• API Integration: Connecting applications with external services to enhance functionality.

2.4 Professional and Career Objectives

The professional and career objectives of the internship were designed to prepare students for
professional environments by developing workplace discipline, technical confidence, and
continuous learning habits. In addition to technical learning, the internship emphasized
professional growth and career readiness through structured learning practices and independent
learning responsibilities.

Students developed the ability to manage learning schedules, complete assignments within
deadlines, and maintain consistency in a self-paced remote learning environment. This helped
improve discipline, responsibility, and adaptability, which are essential qualities in professional
careers.

The internship also provided insight into emerging career opportunities in Artificial Intelligence,
Machine Learning, and AI engineering domains, helping students understand future industry
trends and required skill sets.
Key Professional and Career Objectives:
• Career Readiness: Understanding industry expectations, roles, and responsibilities.
• Technical Confidence: Developing confidence in understanding and working with AI
engineering concepts and tools.
• Interview Preparation: Developing confidence for technical and HR interviews.
• Communication Skills: Improving verbal and written communication.
• Professional Discipline: Developing consistency and responsibility.
• Time Management: Managing tasks and deadlines efficiently.
• Team Collaboration: Learning teamwork and coordination.
• Professional Mindset Development: Building discipline and a growth-oriented attitude.
• Exposure to AI Career Opportunities: Understanding various career paths in technology.

2.5 Project-Specific Objectives

Department of AI&ML, BTI, Bengaluru-35 9

Industry Internship 2025-26
The project-specific objectives of the internship were aimed at providing practical exposure to AI
engineering concepts through real-world AI applications and industry-focused project scenarios.
These objectives focused on strengthening technical understanding, practical implementation
skills, and analytical problem-solving abilities.

The internship program included exposure to AI project workflows across multiple industries
such as healthcare, finance, retail, and manufacturing. Students learned how Artificial
Intelligence can be applied to solve domain-specific challenges through intelligent systems,
machine learning models, and automation techniques.

The internship also emphasized understanding the complete lifecycle of AI system development,
including problem identification, data handling, model development, evaluation, deployment
strategies, and scalability considerations. Through assignments and practical learning modules,
students gained insight into how AI technologies are integrated into modern applications and
business environments.

The program further introduced advanced AI concepts such as Generative AI, Retrieval-
Augmented Generation (RAG), AI agents, multimodal systems, and production-ready AI
pipelines, helping students understand the future direction of AI technologies and applications.

Together, these learning activities provided comprehensive exposure to AI engineering practices

and strengthened students’ understanding of system development, AI workflows, model
evaluation, and real-world implementation strategies. The internship experience also encouraged
innovation, technical exploration, and practical thinking, ensuring overall technical and
professional growth throughout the program.

Internship syllabus and AI Engineer Accelerator Program: Build Real-World AI Projects in

4 Industries (REMOTE)” details referenced from the internship documentation.

Department of AI&ML, BTI, Bengaluru-35 10

Industry Internship 2025-26
Chapter 3

LEARNING EXPERIENCES
The learning experience during the internship at iSpark learning Solutions was comprehensive,
structured, and highly practical in nature. The internship combined theoretical understanding
with hands-on implementation, enabling learners to develop strong foundational and
intermediate-level knowledge in Artificial Intelligence, Machine Learning, Deep Learning, and
AI engineering workflows. The program provided a systematic environment where learning
extended beyond traditional instruction through assignments, quizzes, practical exercises, and
exposure to real-world AI applications.

A major aspect of the internship experience was the emphasis on continuous learning and
practical understanding. Through structured learning modules, guided sessions, self-paced
activities, and industry-focused concepts, students were encouraged to explore emerging
technologies independently and apply their knowledge in practical scenarios. The phased
learning structure enabled gradual progression from AI fundamentals to advanced concepts such
as Generative AI, Natural Language Processing, MLOps, AI agents, and production AI systems.

The internship also focused on developing discipline, adaptability, and consistency through daily
learning activities, assessments, and practical exercises. Students were introduced to modern AI
frameworks and technologies including TensorFlow, Keras, transformer models, and cloud-
based AI systems. This practical exposure significantly improved conceptual understanding,
analytical thinking, and confidence in handling technical challenges.

3.1. Technical Learning

The technical learning during the internship focused on building strong foundational knowledge
and practical understanding of Artificial Intelligence and AI engineering concepts. The
internship at iSpark learning Solutions provided exposure to modern AI technologies, real-world
workflows, and practical implementation methodologies, enabling learners to understand how
theoretical concepts are applied in professional environments.
Key Areas of Technical Learning:

• Artificial Intelligence Fundamentals: Developed understanding of AI, its evolution,

scope, and real-world applications across industries.
• Machine Learning Concepts: Learned supervised and unsupervised learning techniques,
regression, classification, model training, testing, and validation concepts are all in it.

Department of AI&ML, BTI, Bengaluru-35 11

Industry Internship 2025-26

• Deep Learning Fundamentals: Explored neural networks, perceptron, activation

functions, loss functions, and backpropagation techniques.
• Natural Language Processing: Understood text preprocessing, transformer architecture,
BERT, GPT, Large Language Models, and prompt engineering concepts.
• AI-powered recommendation systems: Understood how intelligent systems can analyze
user input and generate personalized suggestions for real-world applications.
• API integration: Learned how to connect applications with external services and APIs to
extend functionality and enable intelligent features.
• Development tools and environments: Gained hands-on experience with tools such as
Android Studio, Google AI Studio, and cloud-based labs used in modern development
workflows.
• Debugging and testing: Improved the ability to identify errors, troubleshoot issues, and
ensure smooth application performance.

3.2. Tools and Technologies

During the internship, several modern AI tools, frameworks, and development technologies were
utilized to support learning, experimentation, and practical implementation. These technologies
provided valuable exposure to industry-standard AI engineering environments and workflows.

TensorFlow:

TensorFlow is an open-source machine learning framework widely used for developing and
deploying Artificial Intelligence and Deep Learning models. During the internship, TensorFlow
was used to understand model creation, neural network implementation, and AI workflow
development. T

Keras:

Keras is a high-level neural network API integrated with TensorFlow that simplifies the process
of building and experimenting with deep learning models. During the internship, Keras was used
for implementing basic classification models and understanding neural network workflows. Its
simplified structure helped in learning model architecture design, training procedures, and
evaluation techniques more effectively.

Google Cloud Labs:

Google Cloud Labs provided hands-on exposure to cloud computing concepts and AI-related
cloud environments. The labs offered guided practical exercises that introduced learners to cloud

Department of AI&ML, BTI, Bengaluru-35 12

Industry Internship 2025-26
infrastructure, scalability concepts, deployment workflows, and resource management. This
helped in understanding how AI systems are managed and deployed in cloud-based professional
environments.

AI Engineering Learning Platform:

The internship utilized a structured online learning platform that organized lessons, quizzes,
assignments, and progress tracking systematically. The platform ensured consistent learning
progression and helped maintain discipline through scheduled modules and assessments. It also
facilitated self-paced learning while allowing continuous monitoring of technical progress.

Machine Learning and AI Technologies:

The internship introduced multiple AI technologies and methodologies, including:
• Supervised and unsupervised learning techniques
• Neural network architectures
• Convolutional Neural Networks (CNNs)
• Recurrent Neural Networks (RNNs) and LSTMs
• Transformer architectures
• Generative AI systems
• Retrieval-Augmented Generation (RAG)
• AI agents and multimodal systems

3.3. Project-Based Learning Experience

The project-based learning experience was one of the most important aspects of the internship
because it encouraged practical understanding and application-oriented learning. Instead of
limiting learning to theoretical concepts, the internship emphasized understanding how AI
technologies are implemented in real-world systems and industry applications. This approach
significantly improved problem-solving skills, analytical thinking, and technical confidence. The
internship program was structured around multiple AI engineering phases that progressively
introduced learners to real-world AI workflows, industry use-cases, and practical AI system
design.

Practical AI Learning Experience

The second phase focused on practical AI engineering concepts such as model evaluation,
feature engineering, overfitting, underfitting, and model optimization techniques. Students were
introduced to neural networks, Computer Vision, NLP systems, and practical AI workflows. This
phase strengthened understanding of how AI models are trained, and improved in real-world
environments.
Department of AI&ML, BTI, Bengaluru-35 13
Industry Internship 2025-26
As the development progressed, the focus shifted to structuring the application and preparing it for
real-time functionality and location-based services. Planning features such as Google Maps
integration, Firebase-based data management, and tourism information retrieval provided a clear
understanding of how modern mobile applications manage data synchronization and navigation
systems. It also introduced concepts related to application architecture, data flow, and interaction
between frontend and backend components.

Another important aspect of this project was understanding scalability and performance
considerations. The application was designed to support multiple users accessing tourism
information and navigation services simultaneously, which required careful planning to ensure
smooth performance and efficient data handling. This highlighted the importance of optimization
and responsive application behavior in Android development.

The project also provided exposure to integrating additional features such as nearby attraction
suggestions, heritage information management, and interactive navigation support. These features
required thinking beyond basic functionality and focusing on enhancing overall user experience.
It demonstrated how value-added features can improve application usability and create a more
impactful digital tourism platform.

Overall Experience

The overall project-based learning experience encouraged independent exploration,

experimentation, and continuous improvement. It helped in developing technical confidence,
practical understanding of AI workflows, and the ability to connect theoretical concepts with
real-world implementation strategies. The experience also strengthened analytical thinking and
prepared learners for future AI engineering projects and professional opportunities.

3.4. Soft Skills and Professional Development

The internship experience was not limited to technical learning but also emphasized the
development of soft skills and professional competencies that are essential for success in
professional environments. The structured learning environment encouraged discipline,
consistency, self-learning, and effective management of responsibilities.

One of the major areas of development was time management and consistency. Since the internship
followed a self-paced and remote learning model, students were required to manage daily lessons,
quizzes, and assignments independently while maintaining regular progress. This improved
discipline, accountability, and the ability to work systematically.

The internship also enhanced communication and comprehension skills through continuous
interaction with technical content, practical exercises, and structured learning activities.
Department of AI&ML, BTI, Bengaluru-35 14
Industry Internship 2025-26
Understanding AI concepts, interpreting technical information, and applying knowledge in
assignments improved analytical communication and technical understanding.

Another important aspect was adaptability and continuous learning. Throughout the internship,
learners were introduced to new tools, frameworks, and advanced AI concepts, which
encouraged flexibility and the ability to adapt to rapidly evolving technologies. This helped in
building confidence while learning unfamiliar concepts and technologies independently.

The internship further contributed toward professional growth by encouraging a growth-oriented

mindset, self-improvement, and continuous skill development. Exposure to emerging AI domains
and industry-focused learning modules helped in understanding future career opportunities in
Artificial Intelligence and AI engineering fields.

3.5. Problem-Solving and Analytical Thinking

The internship played a significant role in strengthening problem-solving and analytical thinking
abilities, which are essential skills in Artificial Intelligence and software development domains.
The learning process emphasized logical thinking, conceptual understanding, and systematic
analysis rather than relying solely on memorization or direct solutions.

One of the key aspects of analytical development was learning how to break down complex
problems into smaller and manageable components. This approach helped in understanding AI
workflows step by step, including data handling, model evaluation, optimization techniques, and
deployment concepts. It also improved understanding of how different system components
interact within AI applications.

The internship also improved decision-making abilities by encouraging evaluation of multiple

approaches before selecting the most suitable solution. Concepts such as model selection, feature
engineering, hyperparameter tuning, and optimization strategies required analytical comparison
and critical thinking to determine effective outcomes.

Another important area of development was debugging and troubleshooting. During practical
exercises and AI model implementation, identifying errors, analyzing system behavior, and
resolving issues systematically helped improve logical reasoning and technical confidence. Over
time, it became easier to identify patterns in errors and approach technical challenges more
efficiently.

The internship further enhanced adaptability and independent thinking by introducing unfamiliar
concepts and advanced technologies progressively. Students were encouraged to explore
solutions independently, improving self-learning capabilities and confidence in handling new

Department of AI&ML, BTI, Bengaluru-35 15

Industry Internship 2025-26
technical domains.

Overall, the internship Overall, the internship provided a strong foundation in analytical thinking,
logical reasoning, technical troubleshooting, and structured problem-solving approaches, all of
which are essential for future professional growth in AI and technology fields.

3.6. Practical Implementation

The internship strongly emphasized practical implementation as an essential part of the learning
process. Instead of focusing only on conceptual knowledge, the program encouraged active
application of AI engineering concepts through assignments, quizzes, hands-on exercises, and
practical AI workflows. This helped bridge the gap between theoretical learning and real-world
technical implementation.
Key Aspects of Practical Implementation:

• Application of theoretical concepts: Concepts learned during sessions were immediately

applied in practical tasks, helping in better understanding and long-term retention of
knowledge.
• Hands-on experience with tools: Actively working with development tools and platforms
improved familiarity and confidence in using industry-standard technologies.
• Learning by doing approach: Instead of passive learning, the focus was on actively
building, testing, and refining solutions, which enhanced engagement and effectiveness.
• Understanding real-world workflows: Practical implementation provided insights into
how tasks are structured, executed, and managed in professional environments.
• Improvement through iteration: Continuous testing and refinement helped in identifying
mistakes, correcting them, and improving the overall quality of work.
• Development of technical confidence: Regular practice reduced hesitation and improved
the ability to handle tasks independently.
• Enhancement of accuracy and efficiency: Repeated implementation helped in reducing
errors and increasing speed and precision in completing tasks.
• Encouragement of experimentation: Exploring different approaches and solutions
allowed for deeper learning and innovation.
• Self-learning and independence: Practical tasks encouraged independent thinking and
problem-solving, reducing reliance on external help.
• Better understanding of system behavior: Implementing concepts provided clarity on
how data flows, how components interact, and how systems respond to user actions.

Department of AI&ML, BTI, Bengaluru-35 16

Industry Internship 2025-26
Chapter 4
LEARNING OUTCOMES
The learning outcomes of the internship represent the knowledge, technical understanding, and
professional competencies gained throughout the entire internship program at iSpark Learning
Solutions. These outcomes reflect how the internship contributed toward both technical growth
and overall professional development through structured learning, practical implementation,
industry-oriented exposure, and continuous skill enhancement.

The internship was designed to ensure that learning extended beyond theoretical understanding
and focused on developing the ability to apply concepts in practical and real-world AI
engineering scenarios. Through exposure to Artificial Intelligence workflows, data pipelines,
model training concepts, backend systems, cloud deployment practices, and modern AI tools,
there was a significant improvement in understanding how production-level AI systems are
designed, implemented, and managed.

The combination of conceptual learning, practical assignments, AI engineering workflows, and

advanced topics such as Retrieval-Augmented Generation (RAG), Generative AI, AI agents,
cloud deployment, and monitoring systems played an important role in strengthening technical
confidence and analytical thinking. The internship also emphasized adaptability, independent
learning, and professional discipline through self-paced modules, quizzes, assignments, and
continuous evaluations.

In addition to technical outcomes, the internship contributed toward developing professional

qualities such as consistency, time management, communication, and a growth-oriented mindset.
Overall, the learning outcomes represent balanced and holistic development, combining AI
engineering knowledge, practical implementation capabilities, and professional readiness for
future academic and career opportunities in the field of Artificial Intelligence and technology.

4.1. Practical Knowledge

One of the most significant outcomes of the internship was the development of strong practical
knowledge in Artificial Intelligence and AI engineering workflows. While academic learning
primarily focuses on theoretical understanding, the internship provided opportunities to
understand how these concepts are implemented in real-world systems and production
environments. This practical exposure helped in gaining deeper understanding of AI systems,
workflows, and modern development practices.

A major realization during the internship was that practical implementation involves multiple

Department of AI&ML, BTI, Bengaluru-35 17

Industry Internship 2025-26
stages including planning, data preparation, model evaluation, deployment considerations, and
system monitoring. Unlike academic examples, real-world AI systems require careful design
decisions, structured workflows, and continuous evaluation to ensure reliability and efficiency.
This helped in developing a broader understanding of how AI systems function beyond isolated
theoretical concepts.
Key Aspects of Practical Knowledge Gained:

• Contextual application of AI concepts: Learned how AI, Machine Learning, and Deep
Learning concepts are applied differently depending on business requirements, system
constraints, and industry use-cases.
• Structured problem-solving approach: Developed the ability to approach tasks in a
step-by-step and organized manner rather than working randomly. Planning before
execution ensured clarity, reduced errors, and improved overall efficiency. This also
helped in maintaining consistency and better time management during implementation.
• Understanding AI system workflows: Gained understanding of AI workflows including
data pipelines, model evaluation, backend integration, deployment processes, and
monitoring systems.
• Integration of multiple technologies: Learned how multiple technologies such as AI
frameworks, cloud environments, APIs, backend systems, and deployment tools work
together within modern AI applications.
• Adaptability during execution: Learned to adjust approaches when faced with
challenges, errors, or changing requirements. Instead of sticking rigidly to one method, it
became important to modify strategies based on the situation. This improved flexibility and
the ability to handle unexpected scenarios effectively.
• Focus on efficiency and optimization: Understood that a solution should not only work
but should also be efficient and user-friendly. This included minimizing errors, improving
performance, and ensuring smooth functionality. It encouraged thinking beyond basic
implementation toward creating better-quality outcomes.

Another important outcome was the development of a system-level perspective, where focus
shifted from individual tasks to understanding how entire AI systems operate. This included
understanding how data flows through pipelines, how models interact with APIs, how backend
systems handle requests, and how deployment environments maintain reliability and scalability.

Practical implementation also improved confidence and independence. Repeated exposure to AI

workflows, system architecture concepts, and implementation-based learning reduced hesitation
and strengthened the ability to approach technical challenges logically and independently.

Overall, the development of practical knowledge helped in transitioning from theoretical

Department of AI&ML, BTI, Bengaluru-35 18
Industry Internship 2025-26
understanding to practical AI engineering capability, creating a strong foundation for future
projects and professional growth.

4.2. Industry Exposure and Work Culture Understanding

The internship provided valuable exposure to industry-oriented AI engineering practices and

professional work culture. It offered insights into how modern technology organizations structure
AI workflows, manage projects, evaluate systems, and maintain reliability in production
environments. This exposure helped in understanding the difference between theoretical learning
and real-world engineering practices where scalability, monitoring, efficiency, and continuous
improvement are essential.

One of the key outcomes was gaining understanding of professional expectations and structured
workflows followed in technology-driven environments. The internship emphasized discipline,
consistency, accountability, and systematic execution of tasks through daily lessons, quizzes,
assignments, and milestone-based progression.

Key Aspects of Industry Exposure and Work Culture:

• Understanding workplace expectations: Gained clarity on what is expected from

individuals in a professional environment, including responsibility, consistency, and
commitment to work. This helped in aligning personal work habits with industry standards.
• Structured workflow and processes: Learned how tasks are organized, assigned, and
executed in a systematic manner. This included following proper steps such as planning,
implementation, and review, ensuring efficiency and clarity in work.
• Importance of deadlines and time management: Understood the significance of
completing tasks within specified timelines. This improved the ability to prioritize work,
manage time effectively, and maintain productivity.
• Professional communication: Developed an understanding of how to communicate ideas
clearly and effectively in a professional environment. This included both verbal and written
communication, which is essential for collaboration and coordination.
• Team collaboration and coordination: Gained insights into how individuals work
together in teams, share responsibilities, and support each other to achieve common goals.
This emphasized the importance of cooperation and mutual understanding.
• Adaptability to work environments: Learned to adjust to different work scenarios, tools,
and expectations. This helped in becoming more flexible and open to learning in dynamic
environments.

The internship also highlighted that AI engineering is not limited to model development alone,
but also includes deployment reliability, system scalability, monitoring, documentation,
Department of AI&ML, BTI, Bengaluru-35 19
Industry Internship 2025-26
evaluation metrics, and user-focused design considerations. This understanding contributed
significantly toward developing a professional mindset and industry-oriented perspective.

4.3. Achievements and Certifications

The internship resulted in several important achievements that reflect technical growth, practical
understanding, and successful completion of industry-oriented AI engineering learning activities.
These achievements demonstrate not only the ability to understand modern Artificial Intelligence
concepts but also the capability to apply them in practical implementation scenarios and real-
world system workflows. They serve as measurable outcomes of the internship, highlighting
development in both technical and professional domains.

Major Achievements:

• Completion of AI Engineering Learning Phases: Successfully completed structured

learning modules covering Artificial Intelligence, Machine Learning, Deep Learning,
Natural Language Processing, Generative AI, MLOps, recommendation systems, and AI
deployment concepts.
• Development of a Capstone AI System: Worked on and developed a comprehensive
AI-based capstone project titled “Retail AI Decision Assistant,” which combined
recommendation systems, demand forecasting, analytics dashboards, and intelligent
decision-support functionalities.
• Development of an AI Automation Side Project: Developed an AI-assisted internship
diary automation system capable of generating internship diary entries using Generative
AI and automated API integration workflows.
• Hands-on Exposure to AI Engineering Tools and Frameworks: Gained practical
experience working with modern AI technologies such as TensorFlow, Keras, Streamlit,
FastAPI concepts, backend APIs, cloud-based systems, and deployment-oriented
workflows.
• Understanding Production AI Systems: Learned concepts related to AI system
architecture, Retrieval-Augmented Generation (RAG), backend integration, data
pipelines, monitoring systems, containerization, and scalable AI infrastructure.
• Successful Completion of Internship Tasks and Assessments: Completed structured
assignments, quizzes, practical exercises, and implementation-based learning activities
within the internship timeline, reflecting consistency, discipline, and commitment toward
continuous learning.

Professional Development Achievements

• Improved analytical thinking and structured problem-solving skills
Department of AI&ML, BTI, Bengaluru-35 20
Industry Internship 2025-26
• Developed technical confidence in understanding AI systems and workflows
• Enhanced independent learning and adaptability to emerging technologies
• Strengthened understanding of professional AI engineering practices
• Maintained steady progress throughout the program, showing dedication and the ability
to achieve set milestones.

4.4. AI Engineering and System Design Learning

One of the most valuable outcomes of the internship was gaining exposure to AI engineering
workflows and production-level system design concepts. The internship introduced structured
methodologies used in real-world AI projects, including problem scoping, architecture planning,
evaluation strategies, deployment systems, and monitoring mechanisms.

A major learning outcome was understanding the importance of properly defining AI problems
before implementation. Concepts such as user requirements, workflow mapping, scope
definition, evaluation criteria, and success metrics highlighted the importance of planning before
model development. This helped in understanding how AI systems are aligned with business
goals and real-world requirements.

The internship also provided understanding of AI system architecture and approach selection.
Different AI implementation strategies such as prompting, Retrieval-Augmented Generation
(RAG), fine-tuning, and agentic systems were explored along with their practical applications
and tradeoffs. This improved understanding of how different AI architectures are selected
depending on scalability, latency, cost, and business requirements.

Another important learning outcome was understanding evaluation metrics and continuous
improvement workflows. Technical metrics, product metrics, and business metrics demonstrated
how AI systems are monitored and improved continuously after deployment. This highlighted
that production AI systems require ongoing evaluation and refinement rather than one-time
implementation.

Overall, these concepts provided a deeper understanding of how production-grade AI systems are
planned, evaluated, and maintained in professional environments.

Tools and Technologies Used:

• Google AI Studio: Google AI Studio was used as a platform to experiment with generative
AI models and understand how different prompts influence output quality. It played a
crucial role in learning prompt engineering techniques, which are essential for obtaining
accurate and context-aware responses.

Department of AI&ML, BTI, Bengaluru-35 21

Industry Internship 2025-26
• Gemini API (Generative AI): The Gemini API was used as the core component for
generating intelligent grocery suggestions and user assistance. It processes user queries and
produces meaningful and personalized responses.
• Programming Environment: The development environment was used to implement the
application logic, user interaction handling, and API integration.
• API Integration Mechanisms: APIs were used to establish communication between the
application and AI services. This involved sending user queries, receiving responses, and
managing data flow efficiently within the system.

Tech Stack:

• Frontend Layer: A simple and intuitive interface was designed to allow users to interact
with the system easily and access grocery-related assistance without difficulty.
• Backend Logic: The backend handled user requests, data processing, and AI
communication efficiently.
• AI Integration Layer: The integration of the Gemini API formed the intelligence layer of
the system, responsible for generating personalized recommendations and responses.
• Data Handling and Storage: User interaction data and grocery-related information were
processed and managed efficiently to improve response quality and usability.

Key Features of the Application:

• AI-powered grocery recommendations.

• Personalized shopping assistance.
• Natural language interaction.
• Smart product suggestions.
• User-friendly interface.
• Efficient grocery management support.
• Context-aware AI responses.

Functional Workflow:
1. The user enters grocery-related queries or requirements.
2. The system processes the user input and analyzes the request.
3. The request is sent to the Gemini API for intelligent response generation.
4. The AI model generates personalized grocery suggestions and recommendations.
5. The generated response is displayed to the user through the application interface.

Outcome and Learning:

This project provided in-depth exposure to AI system design, API integration, and intelligent

Department of AI&ML, BTI, Bengaluru-35 22

Industry Internship 2025-26
application development. It helped in understanding how Generative AI models can be integrated
into real-world applications to improve user convenience and personalization.
The experience also enhanced skills in prompt engineering, AI interaction, and frontend-backend
integration. Moreover, the project demonstrated the importance of usability, efficiency, and
relevance in AI-based systems.

4.5. Main Project: Retail AI Decision Assistant

The capstone project developed during the internship was the “Retail AI Decision Assistant,” an
AI-powered intelligent retail analytics and recommendation system designed to improve product
recommendation, customer personalization, and demand forecasting workflows. In modern retail
environments, organizations often face challenges in understanding customer preferences,
managing inventory efficiently, and predicting product demand accurately. Traditional
recommendation approaches are frequently limited in personalization and adaptability. The
primary objective of this project was to design and study an intelligent AI-driven system capable
of generating personalized recommendations and forecasting product demand using Machine
Learning and AI engineering concepts.

The project was designed around the concept of intelligent retail decision support, where
customer interactions, product behavior, and sales patterns are analyzed to generate meaningful
recommendations and predictive insights. Instead of relying solely on static rule-based systems,
the project utilized AI-driven recommendation workflows and forecasting methodologies to
improve personalization, scalability, and analytical capability.

The system combines recommendation engines, demand forecasting, AI evaluation metrics,

interactive dashboards, and analytics visualization into a unified AI engineering solution. It
demonstrates the practical implementation of Machine Learning workflows, recommendation
architectures, forecasting systems, and AI-driven analytics within real-world retail environments.

Project Description and Objective:

The core objective of the Retail AI Decision Assistant was to build a production-oriented AI
system capable of assisting retail decision-making through intelligent recommendations and
predictive analytics. The project focused on understanding how recommendation systems and
forecasting models can be integrated into business workflows to improve customer experience
and inventory planning.

Department of AI&ML, BTI, Bengaluru-35 23

Industry Internship 2025-26
Another important objective of the project was to understand how AI engineering systems are
designed using structured workflows such as data pipelines, model evaluation, recommendation
ranking, forecasting systems, and analytics monitoring. The project also emphasized system-
level understanding of scalability, modular architecture, and AI deployment-oriented thinking.

The project additionally focused on improving analytical understanding through practical

exposure to collaborative filtering systems, recommendation architectures, user-item similarity
models, and forecasting methodologies.
Tools and Technologies Used:
• Python: Python served as the primary programming language for implementing
recommendation workflows, forecasting logic, data processing, and backend
functionalities.
• Streamlit: Streamlit was used for developing an interactive analytics dashboard and
visualization interface for displaying recommendations, forecasting outputs, and AI
metrics.
• TensorFlow and Keras: TensorFlow and Keras were explored for understanding AI
model workflows, recommendation systems, and Machine Learning experimentation.
• Scikit-learn: Scikit-learn was used for implementing Machine Learning functionalities
such as similarity calculations, regression workflows, evaluation metrics, and
recommendation logic.
• Plotly: Plotly was utilized for interactive data visualization, forecasting graphs, KPI
dashboards, and analytics presentation.
• Pandas and NumPy: Pandas and NumPy were used for data preprocessing,
transformation, analytics handling, and structured data operations.
Key Features of the Application
• AI-powered personalized recommendation system.
• Demand forecasting and predictive analytics.
• Interactive analytics dashboard.
• Recommendation ranking and scoring mechanisms.
• Customer-product similarity analysis.
• KPI monitoring and visualization.
• Intelligent retail decision-support workflows.
Outcome and Learning

The development and study of the Retail AI Decision Assistant provided valuable exposure to
recommendation systems, Machine Learning workflows, forecasting models, analytics

Department of AI&ML, BTI, Bengaluru-35 24

Industry Internship 2025-26
dashboards, and production-oriented AI engineering concepts. It significantly improved on the
understanding of how modern AI systems are designed, evaluated, and integrated into real-world
business applications.

One of the major learning outcomes was understanding how recommendation systems operate
using collaborative filtering, similarity analysis, ranking systems, and personalization workflows.
The project also improved understanding of forecasting methodologies, analytical evaluation,
KPI tracking, and interactive AI visualization systems.

Another important outcome was gaining practical understanding of AI engineering architecture,

including frontend-backend interaction, analytics presentation, modular workflows, and scalable
AI system design. The project additionally strengthened analytical thinking, structured problem-
solving, and system-level understanding of intelligent applications.

Overall, the Retail AI Decision Assistant stands as a comprehensive and impactful capstone
project outcome of the internship. It demonstrates the practical integration of recommendation
systems, Machine Learning, forecasting workflows, and AI-driven analytics into a unified
intelligent retail solution capable of supporting modern business decision-making.

4.6. Professional Development and Career Readiness

The internship significantly contributed toward professional growth and career readiness by
introducing industry-oriented learning practices, portfolio development concepts, documentation
standards, and technical presentation skills. These outcomes helped in understanding how
technical skills are communicated and presented professionally in the AI industry.

One important outcome was learning the significance of documentation and technical
presentation. Concepts related to building demos, writing professional README files, creating
portfolio case studies, and presenting AI projects highlighted the importance of effectively
communicating technical work.

The internship also provided understanding of portfolio building, resume tailoring, LinkedIn
positioning, and interview preparation strategies for AI engineering roles. This helped in
understanding how technical projects, deployment experiences, and AI workflows are presented
professionally to recruiters and employers.

Key Professional Outcomes

• Improved technical communication and documentation skills

• Developed understanding of portfolio presentation and project case studies
• Learned professional AI engineering terminology and workflows

Department of AI&ML, BTI, Bengaluru-35 25

Industry Internship 2025-26
• Improved confidence in explaining AI systems and technical decisions
• Enhanced self-learning, adaptability, and growth-oriented mindset
• Understood industry expectations for AI engineering roles and career opportunities

The internship also strengthened independent learning habits and encouraged continuous
exploration of emerging technologies, which is essential in the rapidly evolving field of Artificial
Intelligence.

4.7. Overall Internship Outcome

The internship at iSpark Learning Solutions resulted in comprehensive growth in terms of

technical understanding, practical implementation capability, and professional readiness. It
provided a structured learning environment where concepts were not only studied theoretically
but also connected with real-world AI engineering workflows and industry practices.

One of the most significant outcomes was the enhancement of technical competence in Artificial
Intelligence, Machine Learning, Deep Learning, backend systems, cloud deployment, and AI
engineering workflows. Through practical learning modules and industry-oriented concepts,
there was noticeable improvement in understanding how modern AI systems are designed,
deployed, monitored, and evaluated.

The internship also strengthened analytical thinking, problem-solving abilities, and system-level
understanding. Exposure to AI workflows, deployment systems, monitoring frameworks, and
evaluation methodologies helped in developing a structured and logical approach toward
handling technical challenges.

In addition to technical growth, the internship improved professional skills such as discipline,
consistency, communication, adaptability, and independent learning. The exposure to portfolio
development, technical documentation, and professional AI engineering practices further
contributed toward career readiness and professional confidence.

Overall, the internship outcome represents balanced and holistic development that combines AI
engineering knowledge, practical understanding, deployment awareness, and professional
growth. It has established a strong foundation for future academic projects, AI engineering
opportunities, and professional development in technology-driven industries.

Learning outcomes and advanced AI engineering concepts referenced from internship learning
modules and final-phase AI engineering documentation.

Department of AI&ML, BTI, Bengaluru-35 26

Industry Internship 2025-26
Chapter 5

CONCLUSION

The internship at iSpark Learning Solutions was a comprehensive and highly enriching learning
experience that contributed significantly toward both technical development and professional
growth. Throughout the internship, the learning process followed a structured and industry-
oriented approach that combined theoretical understanding, practical implementation, AI
engineering workflows, and continuous skill enhancement. The phased learning methodology
helped in gradually progressing from fundamental Artificial Intelligence concepts to advanced
AI engineering systems, deployment workflows, and production-oriented technologies.

A major aspect of the internship was the emphasis on continuous learning through structured
modules, quizzes, assignments, and implementation-oriented activities. Instead of limiting
learning to conceptual understanding alone, the internship encouraged practical exposure to real-
world AI workflows, backend systems, deployment architectures, and modern AI engineering
methodologies. This hands-on approach significantly improved understanding of how AI systems
are designed, evaluated, deployed, and maintained in professional environments.

The internship also provided valuable exposure to modern AI frameworks, cloud technologies,
backend systems, and deployment practices. Working with technologies such as TensorFlow,
Keras, backend APIs, Docker-based deployment systems, monitoring frameworks, and cloud
environments helped in understanding how production-grade AI systems operate in real-world
scenarios. Exposure to concepts such as Retrieval-Augmented Generation (RAG), Generative AI,
AI agents, MLOps, monitoring systems, and scalable infrastructure further broadened technical
understanding and industry awareness.

Another important aspect of the internship was the exposure to AI engineering system design and
professional development concepts. Learning about data pipelines, evaluation systems,
deployment reliability, technical documentation, README writing, demo presentation, portfolio
building, and interview preparation highlighted the importance of communication, presentation,
and structured technical thinking in professional AI engineering environments. These concepts
provided understanding not only of how systems are built, but also how technical work is
effectively presented and maintained in industry settings.

The internship also contributed significantly toward improving analytical thinking, logical
reasoning, and structured problem-solving abilities. Continuous exposure to AI workflows,
evaluation methodologies, debugging concepts, and implementation-based learning encouraged a

Department of AI&ML, BTI, Bengaluru-35 27

Industry Internship 2025-26
systematic approach toward handling technical challenges. This helped in improving confidence
while approaching unfamiliar technologies, understanding system-level workflows, and solving
problems independently through analysis and experimentation.

Another valuable outcome was the development of professional qualities such as consistency,
adaptability, discipline, and independent learning. Since the internship followed a structured and
self-paced learning model, it encouraged maintaining regular progress, managing deadlines, and
continuously improving technical understanding. This helped in developing responsibility, self-
learning habits, and the ability to adapt to rapidly evolving technologies and professional
expectations.

The internship also provided a broader understanding of how modern AI systems extend beyond
model development and include deployment infrastructure, monitoring mechanisms, evaluation
frameworks, backend services, cloud systems, and user-focused design considerations. This
system-level perspective significantly improved understanding of how AI engineering functions
in real-world production environments.

Overall, the internship at iSpark Learning Solutions can be considered a highly valuable and
well-rounded learning journey that successfully integrated technical learning, practical
implementation, AI engineering exposure, and professional development into a cohesive
experience. It strengthened technical competence, practical understanding, confidence, analytical
thinking, and professional readiness while also encouraging continuous learning and adaptability.

The exposure to Artificial Intelligence technologies, modern AI engineering workflows, cloud

systems, deployment practices, professional documentation, and industry-oriented learning
methodologies has established a strong foundation for future academic projects, research
opportunities, and career growth in the field of Artificial Intelligence and technology. The
internship experience has prepared for future technical challenges by providing the knowledge,
practical skills, structured thinking, and professional mindset necessary to continuously learn,
adapt, and grow in a rapidly evolving technological environment.

Department of AI&ML, BTI, Bengaluru-35 28

Industry Internship 2025-26
APPENDIX – A
CODE

1. [Link]
from __future__ import annotations

from dataclasses import dataclass

from pathlib import Path
from typing import Tuple

import numpy as np
import pandas as pd
import [Link] as px
import plotly.graph_objects as go
import streamlit as st
from sklearn.linear_model import LinearRegression
from [Link] import mean_absolute_error
from [Link] import cosine_similarity

DATA_DIR = Path("data")
PRODUCTS_FILE = DATA_DIR / "[Link]"
INTERACTIONS_FILE = DATA_DIR / "[Link]"
SALES_FILE = DATA_DIR / "daily_sales.csv"

@dataclass
class ModelBundle:
products: [Link]
interactions: [Link]
sales: [Link]
user_item: [Link]
item_similarity: [Link]

Department of AI&ML, BTI, Bengaluru-35 29

Industry Internship 2025-26
def inject_css() -> None:
[Link](
"""
<style>
.stApp {
background: linear-gradient(160deg, #0f172a 0%, #1e293b 45%, #111827 100%);
color: #f8fafc;
}
.main-title {
font-size: 2.3rem;
font-weight: 800;
line-height: 1.2;
margin-bottom: .5rem;
}
.subtitle {
color: #cbd5e1;
font-size: 1.02rem;
margin-bottom: 1.2rem;
}
.kpi-card {
background: rgba(15, 23, 42, 0.65);
border: 1px solid rgba(148, 163, 184, 0.25);
border-radius: 14px;
padding: 0.9rem 1rem;
backdrop-filter: blur(4px);
}
.kpi-title {
color: #93c5fd;
font-size: .85rem;
margin-bottom: .15rem;
}
.kpi-value {
font-size: 1.4rem;
font-weight: 700;
}
Department of AI&ML, BTI, Bengaluru-35 30
Industry Internship 2025-26
.section-card {
background: rgba(15, 23, 42, 0.55);
border: 1px solid rgba(148, 163, 184, 0.22);
border-radius: 16px;
padding: 1rem;
margin-bottom: 1rem;
}
div[data-testid="stSidebar"] {
background: linear-gradient(180deg, #111827 0%, #0b1220 100%);
}
</style>
""",
unsafe_allow_html=True,
)

def generate_synthetic_data(
users: int = 160,
products_count: int = 36,
days: int = 140,
seed: int = 42,
) -> Tuple[[Link], [Link], [Link]]:
rng = [Link].default_rng(seed)
DATA_DIR.mkdir(parents=True, exist_ok=True)

categories = ["Healthcare", "Manufacturing", "Retail", "Fintech", "EdTech", "Energy"]

products = []
for pid in range(1, products_count + 1):
category = categories[(pid - 1) % len(categories)]
base_price = [Link](15, 300)
[Link](
{
"product_id": f"P{pid:03d}",
"product_name": f"{category} Product {pid}",
"category": category,
Department of AI&ML, BTI, Bengaluru-35 31
Industry Internship 2025-26
"price": float(base_price),
}
)
products_df = [Link](products)
products_df.to_csv(PRODUCTS_FILE, index=False)

# User-product interactions used for recommendations.

interactions = []
for uid in range(1, users + 1):
user_id = f"U{uid:03d}"
# Each user interacts more with one preferred category.
preferred_category = [Link](categories)
preferred_products = products_df.loc[
products_df["category"] == preferred_category, "product_id"
].tolist()
other_products = products_df.loc[
products_df["category"] != preferred_category, "product_id"
].tolist()

pref_count = [Link](10, 22)

other_count = [Link](4, 11)
chosen = [Link](preferred_products, size=min(pref_count, len(preferred_products)),
replace=False).tolist()
chosen += [Link](other_products, size=min(other_count, len(other_products)),
replace=False).tolist()

for product_id in chosen:

[Link](
{
"user_id": user_id,
"product_id": product_id,
"rating": int([Link](3, 6)),
"interaction_count": int([Link](1, 10)),
"days_ago": int([Link](1, 120)),
}

Department of AI&ML, BTI, Bengaluru-35 32

Industry Internship 2025-26
)

interactions_df = [Link](interactions).drop_duplicates(["user_id", "product_id"])

interactions_df.to_csv(INTERACTIONS_FILE, index=False)

# Daily sales for forecasting.

start_date = [Link]().normalize() - [Link](days=days)
dates = pd.date_range(start_date, periods=days, freq="D")
sales_rows = []
for _, row in products_df.iterrows():
base = [Link](22, 100)
trend = [Link](-0.1, 0.35)
weekly_amp = [Link](4, 20)
noise_scale = [Link](2, 7)
for t, date in enumerate(dates):
weekly = weekly_amp * [Link](2 * [Link] * t / 7)
value = base + trend * t + weekly + [Link](0, noise_scale)
units = max(1, int(round(value)))
sales_rows.append(
{
"date": date,
"product_id": row["product_id"],
"units_sold": units,
}
)

sales_df = [Link](sales_rows)
sales_df.to_csv(SALES_FILE, index=False)
return products_df, interactions_df, sales_df

@st.cache_data(show_spinner=False)
def load_data() -> Tuple[[Link], [Link], [Link]]:
if not PRODUCTS_FILE.exists() or not INTERACTIONS_FILE.exists() or not
SALES_FILE.exists():

Department of AI&ML, BTI, Bengaluru-35 33

Industry Internship 2025-26
return generate_synthetic_data()

products_df = pd.read_csv(PRODUCTS_FILE)
interactions_df = pd.read_csv(INTERACTIONS_FILE)
sales_df = pd.read_csv(SALES_FILE, parse_dates=["date"])
return products_df, interactions_df, sales_df

@st.cache_resource(show_spinner=False)
def build_models() -> ModelBundle:
products, interactions, sales = load_data()
user_item = interactions.pivot_table(
index="user_id",
columns="product_id",
values="rating",
aggfunc="mean",
fill_value=0,
)
item_vectors = user_item.[Link]
sim = cosine_similarity(item_vectors)
item_similarity = [Link](sim, index=user_item.columns, columns=user_item.columns)
return ModelBundle(
products=products,
interactions=interactions,
sales=sales,
user_item=user_item,
item_similarity=item_similarity,
)

def recommend_products(bundle: ModelBundle, user_id: str, top_k: int = 5) -> [Link]:

user_vec = bundle.user_item.loc[user_id]
seen_items = set(user_vec[user_vec > 0].[Link]())
if not seen_items:
pop = (
Department of AI&ML, BTI, Bengaluru-35 34
Industry Internship 2025-26
[Link]("product_id")["interaction_count"]
.sum()
.sort_values(ascending=False)
.head(top_k)
.rename("score")
.reset_index()
)
return [Link]([Link], on="product_id", how="left")

scores = [Link](0.0, index=bundle.user_item.columns)

for item in seen_items:
weight = max(user_vec[item], 1.0)
scores += bundle.item_similarity[item] * weight

for item in seen_items:

scores[item] = -1

top_items = scores.sort_values(ascending=False).head(top_k).rename("score").reset_index()
top_items = top_items.rename(columns={"index": "product_id"})
result = top_items.merge([Link], on="product_id", how="left")
result["score"] = result["score"].round(3)
return result

def evaluate_recommender(bundle: ModelBundle, k: int = 5) -> float:

df = [Link]()
df["days_ago"] = pd.to_numeric(df["days_ago"], errors="coerce").fillna(999)
df = df.sort_values(["user_id", "days_ago"])
if [Link]:
return 0.0

test_rows = [Link]("user_id").head(1)
train_df = [Link](test_rows.index)
if train_df.empty:
return 0.0
Department of AI&ML, BTI, Bengaluru-35 35
Industry Internship 2025-26

train_matrix = train_df.pivot_table(
index="user_id",
columns="product_id",
values="rating",
aggfunc="mean",
fill_value=0,
)
all_items = sorted(df["product_id"].unique())
train_matrix = train_matrix.reindex(columns=all_items, fill_value=0)
sim = cosine_similarity(train_matrix.[Link])
item_similarity = [Link](sim, index=train_matrix.columns,
columns=train_matrix.columns)

hits = 0
users_evaluated = 0
for _, row in test_rows.iterrows():
user_id = row["user_id"]
true_item = row["product_id"]
if user_id not in train_matrix.index:
continue

user_vec = train_matrix.loc[user_id]
seen_items = set(user_vec[user_vec > 0].[Link]())
scores = [Link](0.0, index=train_matrix.columns)
for item in seen_items:
scores += item_similarity[item] * max(user_vec[item], 1.0)
for item in seen_items:
scores[item] = -1
recs = scores.sort_values(ascending=False).head(k).[Link]()
hits += int(true_item in recs)
users_evaluated += 1

return hits / max(users_evaluated, 1)

Department of AI&ML, BTI, Bengaluru-35 36

Industry Internship 2025-26

def forecast_series(series: [Link], horizon: int = 14) -> Tuple[[Link], [Link]]:

# A tiny, robust model: trend + weekly seasonality terms.
y = [Link](float)
t = [Link](len(y))
X = np.column_stack(
[
t,
[Link](2 * [Link] * t / 7),
[Link](2 * [Link] * t / 7),
]
)
model = LinearRegression()
[Link](X, y)

future_t = [Link](len(y), len(y) + horizon)

X_future = np.column_stack(
[
future_t,
[Link](2 * [Link] * future_t / 7),
[Link](2 * [Link] * future_t / 7),
]
)
preds = [Link](X_future)
preds = [Link](preds, 0).round(0)

history = [Link]({"index": [Link], "units_sold": y})

forecast = [Link]({"index": pd.date_range([Link][-1] + [Link](days=1),
periods=horizon), "forecast": preds})
return history, forecast

def evaluate_forecast(bundle: ModelBundle) -> float:

all_mae = []
for product_id in [Link]["product_id"].head(14):

Department of AI&ML, BTI, Bengaluru-35 37

Industry Internship 2025-26
s=(
[Link][[Link]["product_id"] == product_id]
.sort_values("date")
.set_index("date")["units_sold"]
)
if len(s) < 30:
continue
train = [Link][:-14]
test = [Link][-14:]
_, pred_df = forecast_series(train, horizon=14)
mae = mean_absolute_error([Link], pred_df["forecast"].values)
all_mae.append(mae)
if not all_mae:
return 0.0
return float([Link](all_mae))

def kpi_card(title: str, value: str) -> None:

[Link](
f"""
<div class="kpi-card">
<div class="kpi-title">{title}</div>
<div class="kpi-value">{value}</div>
</div>
""",
unsafe_allow_html=True,
)

def render_app() -> None:

st.set_page_config(page_title="Retail AI Decision Assistant",
page_icon=":chart_with_upwards_trend:", layout="wide")
inject_css()

bundle = build_models()

Department of AI&ML, BTI, Bengaluru-35 38

Industry Internship 2025-26

users = sorted(bundle.user_item.[Link]())
products = [Link].sort_values("product_name")

with [Link]:
[Link]("Control Panel")
user_id = [Link]("Select Customer", users, index=0)
product_id = [Link]("Select Product", products["product_id"].tolist(), index=0)
horizon = [Link]("Forecast Horizon (days)", min_value=7, max_value=30, value=14)
top_k = [Link]("Top Recommendations", min_value=3, max_value=10, value=5)
[Link]("Interactive controls")

[Link]('<div class="main-title">Retail AI Decision Assistant</div>',

unsafe_allow_html=True)
[Link](
'<div class="subtitle">Unified analytics for personalized recommendations and demand
forecasting.</div>',
unsafe_allow_html=True,
)

hit_rate = evaluate_recommender(bundle, k=5)

forecast_mae = evaluate_forecast(bundle)
total_units = int([Link]["units_sold"].sum())
avg_daily_units = float([Link]("date")["units_sold"].sum().mean())

kpi_cols = [Link](4)
with kpi_cols[0]:
kpi_card("Recommender Hit Rate@5", f"{hit_rate:.1%}")
with kpi_cols[1]:
kpi_card("Forecast MAE", f"{forecast_mae:.2f}")
with kpi_cols[2]:
kpi_card("Total Units in Dataset", f"{total_units:,}")
with kpi_cols[3]:
kpi_card("Avg Units / Day", f"{avg_daily_units:.0f}")

Department of AI&ML, BTI, Bengaluru-35 39

Industry Internship 2025-26
tab1, tab2, tab3, tab4 = [Link](
["Overview", "Recommendations", "Demand Forecast", "System"]
)

with tab1:
[Link]('<div class="section-card">', unsafe_allow_html=True)
by_category = (
[Link]([Link], on="product_id", how="left")
.groupby("category", as_index=False)["units_sold"]
.sum()
.sort_values("units_sold", ascending=False)
)
fig = [Link](
by_category,
x="category",
y="units_sold",
color="category",
title="Total Sales by Category",
template="plotly_dark",
)
st.plotly_chart(fig, use_container_width=True)
[Link]("</div>", unsafe_allow_html=True)

with tab2:
recs = recommend_products(bundle, user_id=user_id, top_k=top_k)
user_history = (
[Link][[Link]["user_id"] == user_id]
.merge([Link], on="product_id", how="left")
.sort_values("interaction_count", ascending=False)
)

c1, c2 = [Link]([1, 1])

with c1:
[Link]('<div class="section-card">', unsafe_allow_html=True)
[Link](f"Top {top_k} Recommendations for {user_id}")
Department of AI&ML, BTI, Bengaluru-35 40
Industry Internship 2025-26
[Link](
recs[["product_id", "product_name", "category", "price", "score"]],
use_container_width=True,
hide_index=True,
)
chart = [Link](
recs.sort_values("score", ascending=True),
x="score",
y="product_name",
color="category",
orientation="h",
template="plotly_dark",
title="Recommendation Scores",
)
st.plotly_chart(chart, use_container_width=True)
[Link]("</div>", unsafe_allow_html=True)

with c2:
[Link]('<div class="section-card">', unsafe_allow_html=True)
[Link](f"{user_id} Interaction History")
[Link](
user_history[["product_id", "product_name", "category", "rating",
"interaction_count"]].head(12),
use_container_width=True,
hide_index=True,
)
[Link](
"""
The model recommends products that are similar to items this user rated highly.
It uses item-to-item cosine similarity on the user-product rating matrix.
"""
)
[Link]("</div>", unsafe_allow_html=True)

with tab3:

Department of AI&ML, BTI, Bengaluru-35 41

Industry Internship 2025-26
product_series = (
[Link][[Link]["product_id"] == product_id]
.sort_values("date")
.set_index("date")["units_sold"]
)
history_df, forecast_df = forecast_series(product_series, horizon=horizon)
product_meta = [Link][[Link]["product_id"] == product_id].iloc[0]

[Link]('<div class="section-card">', unsafe_allow_html=True)

[Link](f"Demand Forecast: {product_meta['product_name']}
({product_meta['category']})")

line = [Link]()
line.add_trace(
[Link](
x=history_df["index"],
y=history_df["units_sold"],
mode="lines",
name="Historical Sales",
line={"color": "#38bdf8", "width": 2.2},
)
)
line.add_trace(
[Link](
x=forecast_df["index"],
y=forecast_df["forecast"],
mode="lines+markers",
name="Forecast",
line={"color": "#22c55e", "width": 2.6},
)
)
line.update_layout(
template="plotly_dark",
xaxis_title="Date",
yaxis_title="Units Sold",

Department of AI&ML, BTI, Bengaluru-35 42

Industry Internship 2025-26
height=460,
)
st.plotly_chart(line, use_container_width=True)

expected_next_week = int(forecast_df.head(7)["forecast"].sum())
[Link](
f"Expected units for next 7 days: **{expected_next_week}**. "
"Use this as a smart inventory planning signal."
)
[Link]("</div>", unsafe_allow_html=True)

with tab4:
[Link]('<div class="section-card">', unsafe_allow_html=True)
[Link]("System Architecture")
[Link](
"""
1. **Data Layer**: synthetic retail products, user interactions, and daily sales.
2. **Recommendation Engine**: item-based collaborative filtering with cosine similarity.
3. **Forecast Engine**: linear regression with trend + weekly seasonality.
4. **Presentation Layer**: Streamlit + Plotly interactive dashboard.
"""
)
[Link]("Model Outputs")
[Link](
"""
- Customer-level ranked product recommendations with relevance scores.
- Product-level multi-day demand forecast for planning and replenishment.
- Live quality indicators: Hit Rate@5 and MAE.
"""
)
[Link]("Technical Notes")
[Link](
"""
- Recommendation uses item-to-item similarity on interaction signals.
- Forecast uses trend and weekly seasonality features with linear regression.
Department of AI&ML, BTI, Bengaluru-35 43
Industry Internship 2025-26
- The pipeline updates automatically with control changes.
"""
)
[Link]("</div>", unsafe_allow_html=True)

if __name__ == "__main__":
render_app()

Department of AI&ML, BTI, Bengaluru-35 44

 Industry Internship 2025-26
APPENDIX – B
SNAPSHOTS
Main Project:
Recommendation Engine
Forecast Chart Forecast Page

Overview Chart Overview page

Department of AI&ML, BTI, Bengaluru-35 45

Industry Internship 2025-26

Recommendation Scores Recommendations Page

Department of AI&ML, BTI, Bengaluru-35 46