UpSkill India Challenge
Hackathon Organized by [HCL GUVI]
In association with Techfest IIT Bombay
Overview
The UpSkill India Challenge is a national-level hackathon organized in association with
HCL GUVI under its flagship initiative Mission Upskill India. The challenge aims to ignite
innovation in educational technology by harnessing the power of Artificial Intelligence,
empowering participants to design transformative AI-driven solutions that redefine the future
of learning and skill development.
The goal is to empower learners, teachers, and institutions by developing intelligent,
scalable, and impactful AI applications that redefine how education is delivered,
personalized, and evaluated.
This challenge will bring together the brightest minds in AI, software engineering, and
education technology to solve real-world challenges that directly enhance the learning
ecosystem.
Aim
To build AI-powered educational systems that enhance engagement, assessment, and
accessibility through innovation and real-world applicability.
Hackathon Structure
The hackathon will be conducted in three major rounds:
Round 1 – Problem Selection and Presentation:
Objective:
Teams choose one of the provided problem statements, research deeply, and present their
understanding, technical feasibility, and proposed solution.
Deliverable:
A video presentation (≤5 minutes) covering:
● Problem understanding
● Proposed approach and architecture
● Key innovation or differentiator
● Technical feasibility and scalability
● Team composition and skill roles
● Preliminary roadmap (tech stack, milestones)
Evaluation Criteria:
Criterion Description Weightage
Problem Understanding Depth of problem insight and clarity 25%
Innovation Uniqueness and creativity of proposed idea 25%
Feasibility Practicality, scalability, and technology 25%
stack
Presentation Clarity, structure, and communication 15%
Team Strength Role clarity, collaboration potential 10%
Output:
Top teams shortlisted for Round 2.
Submission Process:
You must submit your presentation either to jaya@[Link] or through the Google
Form that will be shared just before the submission deadline on the email ID used during
registration.
Please keep checking your inbox regularly for updates and submission details.
Joining the WhatsApp group is mandatory for submission announcements and important
updates — the link is available on the competition page of the Techfest website.
Round 2 – Development Phase (Prototype Building)
Objective:
Teams begin implementing their solution, aiming for 60–80% completion with a functional
prototype demonstrating the core AI and application logic.
Duration:
3 weeks (Remote with mentorship)
Key Milestones:
● Week 1: Setup, architecture finalization, and backend scaffolding
● Week 2: Core AI model integration / API development
● Week 3: Frontend and workflow integration
● Week 4: Feature refinement, testing, and documentation
Evaluation Criteria:
Criterion Description Weightage
Functional Prototype Core AI logic and main workflow 30%
implemented
Technical Depth Architecture, model choice, optimization 25%
Usability User experience and functionality 15%
Progress & Version control, code quality, reports 15%
Documentation
Impact Potential Scalability and relevance to education 15%
Output:
Teams achieving 60–80% completion qualify for the final sprint at IIT Bombay.
Round 3 – Onsite Grand Finale (IIT Bombay)
Objective:
A 10-hour live coding and presentation sprint at IIT Bombay. Teams finalize, polish, and
present their full-fledged applications.
Activities:
● Complete feature integration
● Final testing and debugging
● Prepare live demo and pitch
● Present to judges (industry experts and academia)
Evaluation Criteria:
Criterion Description Weightage
Functionality Completeness and reliability of the solution 25%
AI Integration Depth, originality, and performance of AI 25%
model
Innovation Uniqueness and problem-solving creativity 20%
Educational Impact Real-world usability and learning benefit 15%
Scalability & Tech scalability, deployment plan, demo 15%
Presentation quality
Final Presentation (25–30 mins total):
● Product Overview: Problem, solution, users, and innovation (5 mins)
● Live Demo: Working prototype demonstration (10 mins)
● Technical Deep Dive: Architecture, AI model, scalability (5 mins)
● Impact & Sustainability: Educational outcomes, future roadmap (5 mins)
Q&A: Judges’ interaction (5 mins)
Awards: Winners will be declared post-presentation.
Problem Statements
1. AI Quiz Portal (Platform Replication(like Kahoot) with AI Quiz
Generation)
Goal:
Design and develop an online quiz platform that enables real-time quiz participation,
live leaderboard tracking, and one-click AI-based quiz creation for learners, educators,
and organizations.
Description:
In today’s fast-paced digital learning environment, quizzes play a vital role in engagement,
assessment, and skill evaluation. However, existing quiz tools often lack interactivity,
automation, and scalability. This challenge focuses on building a next-generation quiz
platform that blends AI intelligence, real-time engagement, and seamless user
experience.
Requirements:
● Host live multiplayer quizzes
● Real-time scoring and leaderboard updates
● Admin dashboard for quiz management
● AI feature: Auto-generate quizzes from text, PDFs, or topics
● Scalable backend (microservices preferred)
● Responsive and smooth frontend (React preferred)
Bonus:
● Analytics dashboard
● Adaptive quiz difficulty using AI
● Integration with speech-based questions
Expected Outcome:
A fully functional prototype that allows users to:
● Create quizzes using AI in seconds
● Host and join real-time sessions
● Compete on a live leaderboard
● Analyze performance post-quiz
Impact:
Such a platform can revolutionize edtech engagement, corporate learning, and
hackathon/game-based learning environments by merging AI automation with real-time
interaction.
2. Mentor Scoring AI
Goal:
Develop an AI-driven system that can evaluate the teaching quality and effectiveness
of mentors or instructors based on their recorded video sessions.
Description:
In large-scale education and training ecosystems, assessing the quality of teaching
delivery across numerous mentors is a major challenge. Manual evaluation is subjective,
time-consuming, and inconsistent. This problem statement focuses on creating an
AI-powered evaluation framework that can analyze recorded mentor videos to assess
key teaching metrics such as clarity, engagement, tone, pacing, and content delivery.
Requirements:
● Accept video input (recordings of teaching sessions)
● Analyze communication clarity, engagement, technical correctness,
confidence, and interaction quality
● Provide a teaching score with parameter-wise breakdown
● Use multimodal analysis (audio, visual, text if available)
Bonus:
● Bias mitigation
● Comparative dashboards (top mentors, weak areas)
● Explainable AI outputs
Expected Outcome:
A prototype system that can autonomously assess mentor performance from recorded
sessions, generate AI-based evaluation reports, and help institutions enhance training
quality and consistency across their educator network.
Impact:
This innovation can transform educator evaluation and upskilling by offering an
objective, scalable, and data-driven assessment mechanism—empowering institutions to
improve teaching quality and learners’ overall experience.
Evaluation Metrics:
Engagement (20%), Communication (20%), Technical Depth (30%), Clarity (20%),
Interaction (10%)
3. AI Mock Interview Platform
Goal:
Build an AI-powered interviewer application that can conduct mock interviews in real
time, evaluate candidates’ responses, and provide instant, personalized feedback to help
them improve.
Description:
Preparing for interviews can be stressful, and human-led mock interviews are often limited
by availability, bias, and scalability. This challenge aims to create an AI-driven mock
interview platform that simulates realistic, domain-specific interviews — offering candidates
a chance to practice, analyze, and enhance their communication, confidence, and
technical knowledge through real-time interaction with an AI interviewer.
Requirements:
1. Real-time audio-based conversation with AI
[Link]-specific interview modes (e.g., ML Engineer, Data Analyst, etc.)
3. Evaluation feedback:
● Communication
● Technical correctness
● Confidence and fluency
4. Auto-generated improvement plan and suggested learning materials
Bonus:
● Support for voice and video interviews
● Integration with resume parsing and personalized feedback
● Multi-round interview simulation
Expected Outcome:
A working prototype where users can attend AI-led interviews, receive real-time
evaluation and scoring, and gain actionable feedback to enhance their performance.
Impact:
The solution empowers students, job seekers, and professionals to practice interview
skills anytime, anywhere, making AI-assisted career readiness accessible, scalable, and
personalized.
Eligibility & Team Composition
● Open to all students with valid institutional ID
● Teams can include members from different colleges
● Maximum team size: 4 members
● Cross-domain teams (AI + frontend + backend + design) are encouraged
Competition Timeline
Date Event
12 November 2025 Registration Deadline
12–15 November 2025 Round 1: Idea Presentation
17 November 2025 Shortlisting of Round 1 Teams
18 November -11 December Round 2: Development & Prototype Submission
2025
13 December 2025 Finalists Announcement
21 December 2025 Round 3: Onsite Finale at IIT Bombay [10 hours Live
Sprint]
22 December 2025 Award Ceremony
Submission Process:
You must submit your presentation either to jaya@[Link] or through the Google
Form that will be shared just before the submission deadline on the email ID used during
registration.
Please keep checking your inbox regularly for updates and submission details.
Joining the WhatsApp group is mandatory for submission announcements and important
updates — the link is available on the competition page of the Techfest website.
Rules & Regulations
1. Each team must register before the deadline.
2. Only original work is allowed — plagiarism will result in disqualification.
3. Submissions after deadlines will not be considered.
4. All code and documentation must be uploaded to the provided repository link.
5. Judges’ decisions are final and binding.
6. Teams must maintain version control (GitHub link mandatory).
Deliverables Summary
Round Deliverable Format
1 5-minute problem understanding and proposal MP4 / YouTube link
video
2 Working prototype (60–80%) + documentation GitHub link + Report
3 Final solution demo + pitch deck Live presentation