Python

Python Core Fundamentals

 Python syntax, indentation, code style (PEP8)
 Variables, data types, type casting
 Input/output, comments, docstrings
 Control flow: if / else, loops (for, while)
 Functions, arguments, return values
 Lambda functions
 List / dict / set comprehensions

Built-in Data Structures (Must-Master)

 Lists
 Tuples
 Sets
 Dictionaries
 Strings (immutability, slicing, formatting)

Pythonic Thinking
 Iterators and generators
 zip, enumerate, map, filter, reduce
 Shallow vs deep copy
 Mutability vs immutability
 *args and **kwargs

Object-Oriented Programming (OOP)

 Classes and objects
 Constructors (__init__)
 Instance vs class variables
 Methods and dunder methods
 Inheritance and method overriding
 Polymorphism
 Encapsulation & abstraction
 Dataclasses (@dataclass)

Error Handling & Debugging

 Exceptions (try / except / finally)
 Custom exceptions
 Common runtime errors

Modules, Packages & Environments

 Import system (import, from, as)
 Creating modules and packages
 Virtual environments (venv, conda)
 Dependency management (pip, [Link], poetry)
 Understanding __main__

File Handling & Serialization

 Reading/writing text files
 CSV, JSON handling
 Pickle (pros & cons)
 Working with directories (os, pathlib)
 Logging to files

Numerical Computing (AI Foundation)

 NumPy
 Arrays, shapes, dtypes
 Broadcasting
 Vectorized operations
 Linear algebra basics
 Random sampling
 Data Analysis & Manipulation
 Pandas
 Series and DataFrames
 Indexing & filtering
 GroupBy & aggregation
 Missing values handling
 Merging & joining
 Time-series basics

Data Visualization (Essential)

 Matplotlib
 Seaborn
 Plot types for EDA

Standard Library (High-ROI Parts)

 os, sys
 pathlib
 datetime, time
 math, random
 logging
 Asynchronous & Parallel Python (Modern AI Apps)
 Multithreading vs multiprocessing
 async / await
 Async I/O basics
 When to use each

Other Python Libraries

 FastAPI
 Pydantic

Production-Ready Python Mindset

 Writing clean, readable code
 Modular design
 Logging & monitoring mindset
 Reading other people’s code
[Link]

SQL
SQL Fundamentals
 What SQL is & why databases exist
 Relational database concepts
 Tables, rows, columns
 Primary key vs foreign key
 Basic SQL syntax & formatting
 Data Manipulation (DML)
 INSERT
 UPDATE
 DELETE
 UPSERT (ON CONFLICT)
 Transactions (BEGIN, COMMIT, ROLLBACK)

Table Design & Schema Basics

 Creating tables (CREATE TABLE)
 Constraints:
 PRIMARY KEY
 FOREIGN KEY
 UNIQUE
 NOT NULL
 CHECK
 Indexes (conceptual + practical)
 Normalization basics (1NF, 2NF, 3NF)
 Denormalization (when & why)

Core Querying (Absolute Must)

 SELECT
 FROM
 WHERE
 DISTINCT
 ORDER BY
 LIMIT / OFFSET
 Aliases (AS)

Filtering & Conditions

 Comparison operators (=, !=, <, >, BETWEEN)
 Logical operators (AND, OR, NOT)
 IN, NOT IN
 LIKE, ILIKE
 IS NULL, IS NOT NULL
 Case expressions (CASE WHEN)

Aggregations & Grouping

 Aggregate functions:
 COUNT
 SUM
 AVG
 MIN, MAX
 GROUP BY
 HAVING
 Difference between WHERE and HAVING
 Joins (Critical for Real Data)
 INNER JOIN
 LEFT JOIN
 RIGHT JOIN
 FULL OUTER JOIN
 CROSS JOIN
 Join conditions & join keys
 Self joins
 Common join pitfalls (duplicate rows)

Subqueries & CTEs

 Subqueries in SELECT, FROM, WHERE
 Correlated subqueries
 WITH (Common Table Expressions)
 Recursive CTEs (conceptual)
 When to use CTEs vs subqueries

Window Functions (AI / Analytics Superpower)

 OVER() clause
 PARTITION BY
 ORDER BY in windows
 Ranking functions:
 ROW_NUMBER
 RANK
 DENSE_RANK
 Aggregate window functions
 Lag/lead (LAG, LEAD)
 Running totals & moving averages
 Data Cleaning & Transformation
 String functions (LOWER, UPPER, TRIM, SUBSTRING)
 Date/time functions
 Type casting (CAST, ::)
 Handling missing values
 Conditional transformations
 Normalizing raw data using SQL

Performance & Optimization (High ROI)

 Indexes and how they work
 Query execution plans (EXPLAIN)
 Common slow query patterns
 Filtering early vs late
 Avoiding unnecessary joins
 When SQL beats Pandas

Analytical Thinking with SQL

 Funnel analysis
 Cohort analysis
 Time-based analysis
 User behavior analysis
 Feature extraction using SQL
 Creating ML-ready datasets

SQL Dialects Awareness

 PostgreSQL
 MySQL
 SQLite
 BigQuery / Snowflake (conceptual)
 ANSI SQL vs dialect differences

SQL + Python Integration

 Running SQL from Python
 sqlite3, psycopg2, sqlalchemy
 Pandas + SQL workflows
 Writing parameterized queries
 Avoiding SQL injection

Databases Every AI Engineer Should Touch

 PostgreSQL
 SQLite
 MySQL
 Cloud warehouses (BigQuery / Redshift / Snowflake – awareness)

Real-World SQL Skills

 Writing readable SQL
 Breaking complex queries into steps
 Commenting queries
 Versioning SQL
 Debugging wrong results
[Link]

Mathematics
Linear Algebra (The Backbone of ML)
 Scalars, vectors, matrices, tensors
 Vector and matrix operations
 Dot product & geometric interpretation
 Matrix multiplication
 Identity, transpose, inverse
 Rank of a matrix
 Systems of linear equations
 Linear transformations
 Column space & null space (intuition level)
 Eigenvalues & eigenvectors (intuition + usage)
 Orthogonality
 Norms & distances (L1, L2)
 Calculus (Learning Happens Here)
Differential Calculus
 Functions and graphs
 Limits (intuition only)
 Derivatives and gradients
 Partial derivatives
 Chain rule (very important)
 Gradient as direction of steepest descent
 Local vs global minima
 Convex vs non-convex functions

Optimization Concepts
 Gradient Descent
 Learning rate intuition
 Cost / loss functions
 Saddle points
 Vanishing & exploding gradients (intuition)

Probability Theory (Uncertainty Handling)

 Random variables
 Discrete vs continuous variables
 Probability distributions
 PMF, PDF, CDF
 Expectation & variance
 Common distributions:
 Bernoulli
 Binomial
 Normal
 Uniform
 Poisson
 Independence & conditional probability
 Bayes’ theorem
 Likelihood vs probability

Statistics (From Data to Decisions)

 Population vs sample
 Measures of central tendency
 Measures of dispersion
 Covariance
 Correlation
 Bias & variance
 Sampling techniques
 Central Limit Theorem (intuition)
 Law of Large Numbers
 Outliers & robustness

Information Theory (Modern ML Intuition)

 Entropy
 Cross-entropy
 KL divergence
 Information gain
 Why cross-entropy is used in classification

Geometry & Distance Measures

 Euclidean distance
 Manhattan distance
 Cosine similarity
 Angle between vectors
 High-dimensional intuition (curse of dimensionality)

Probability + Linear Algebra in ML

 Multivariate distributions
 Gaussian distribution in higher dimensions
 Covariance matrix interpretation
 Mahalanobis distance

Matrix Factorization & Decomposition (High ROI)

 Eigen decomposition
 Singular Value Decomposition (SVD)
 PCA math intuition
 Dimensionality reduction rationale

Software Essentials
Version Control
 Git fundamentals
 Branching and merging
 Pull requests
 Code reviews
 Merge conflict resolution
 GitHub workflows

Development Environment
 Virtual environments
 Dependency management
 Environment variables
 Project structure
 Reproducible setups

Working with APIs & Services

 REST APIs
 HTTP methods & status codes
 JSON handling
 Authentication mechanisms
 Error handling
 Rate limiting
Free Resources
 videoGit & Github by CampusX
 videoVirtual Environments
 videoUV in Python
 videoWhat is API

Cloud Fundamentals
 Cloud Fundamentals

Cloud Basics
 What cloud computing is
 Cloud service models (IaaS, PaaS, SaaS)
 Public, private, hybrid cloud
 Regions, availability zones
 Shared responsibility model

Core Cloud Services

 Compute services
 Storage services
 Networking services
 Managed databases
 Managed AI/ML services

Compute Fundamentals
 Virtual machines
 Containers (conceptual)
 Auto-scaling
 Load balancing
 Serverless computing

Storage Fundamentals
 Object storage
 Block storage
 File storage

Networking Basics
 Virtual networks (VPC/VNet)
 Subnets
 IP addressing
 Firewalls and security groups
 DNS
 Load balancers

Databases in the Cloud

 Relational databases
 NoSQL databases
 Managed database services
 Backups and replication

Identity & Security

 Identity and access management (IAM)
 Roles and policies
 Authentication vs authorization
 Secrets management
 Network security basics

Containers & Orchestration (Awareness)

 Docker fundamentals
 Container images
 Container registries

Cloud for AI Workloads

 GPU and accelerator instances
 Batch vs real-time inference
 Model deployment patterns
 Data storage for ML pipelines

Cost & Resource Management

 Pay-as-you-go pricing
 Cost estimation
 Resource tagging
 Budgeting and cost monitoring

DevOps & Automation Basics

 Infrastructure as code (conceptual)
 CI/CD pipelines (conceptual)
 Environment separation (dev, staging, prod)

Cloud Providers Awareness

 AWS
 Google Cloud Platform (GCP)
 Microsoft Azure
 AI Coding
AI-Assisted Development Basics
 What AI coding assistants are
 Code generation vs code completion
 Prompting for code
 Understanding AI-generated code

Prompting for Code

 Writing clear coding prompts
 Providing context and constraints
 Iterative prompting
 Asking for explanations and refactors

Reading & Validating AI Code

 Understanding unfamiliar code
 Identifying logical errors
 Identifying security issues
 Validating outputs
 Testing AI-generated code
 Refactoring with AI
 Code simplification
 Improving readability
 Performance optimization
 Modularization
 Converting scripts into packages

Debugging with AI
 Explaining errors and stack traces
 Hypothesis-driven debugging
 Step-by-step reasoning
 Fixing edge cases

AI for Learning Codebases

 Summarizing large codebases
 Explaining functions and modules
 Generating architecture overviews
 Navigating unfamiliar repositories
 AI for Data & ML Code
 Writing data preprocessing code
 Feature engineering code
 Model training pipelines
 Evaluation code
 Experiment scripts

AI for Automation
 Script generation
 Workflow automation
 Data pipeline scripts
 DevOps-related scripts

AI Coding Tools Awareness

 IDE-based AI assistants
 Chat-based coding assistants
 CLI-based AI tools
 Code review assistants
 Limitations of AI Coding
 Hallucinated APIs
 Outdated patterns
 Overconfidence in incorrect code
 Dependency mismatches

Machine Learning Fundamentals

What Machine Learning Really Is
 What is Machine Learning (vs traditional programming)
 Why ML works: pattern discovery from data
 ML vs Statistics vs Rules-based systems
 When not to use ML

Types of Machine Learning

 Supervised learning
 Unsupervised learning
 Semi-supervised learning
 Self-supervised learning (conceptual intro)
 Reinforcement learning (high-level only)
 Batch learning vs online learning
 ML Problem Formulation
 Regression vs classification
 Binary vs multi-class vs multi-label
 Clustering vs density estimation
 Anomaly / outlier detection
 Ranking & recommendation problems

Train / Validation / Test Split

 Why splitting data matters
 Train vs validation vs test sets
 Data leakage (very important)
 Hold-out validation
 Cross-validation (conceptual, not math-heavy)
 When test data should be touched (almost never)

Model Concept (Very Important)

 What is a model?
 Parameters vs hyperparameters
 Hypothesis space intuition
 Decision boundaries (visual intuition)
 Linear vs non-linear models (conceptual)
 Loss Functions (Optimization Target)
 What is a loss function?
 Loss vs error vs metric
 Regression losses (MAE, MSE – intuition)
 Classification losses (log loss – intuition)
 Why models minimize loss, not accuracy
 Average loss over dataset

Evaluation Metrics (Conceptual)

 Regression metrics: MAE, MSE, RMSE, R² (intuition)
 Classification metrics:
 Accuracy
 Precision, recall
 F1-score
 Confusion matrix intuition
 Why accuracy is often a bad metric
 Choosing the right metric for the problem

 Bias–Variance Tradeoff
 Bias vs variance intuition
 Underfitting vs overfitting
 Model complexity vs data size
 How bias–variance shows up in practice
 Why this tradeoff never disappears
 End-to-End ML Workflow
 Problem definition
 Data collection
 Data understanding
 Train–validate loop
 Model selection
 Final evaluation
 Deployment awareness (high-level)

Common ML Pitfalls (Must-Know)

 Data leakage
 Improper evaluation
 Ignoring baseline models
 Blindly trusting metrics

Free Resources
 video100 Days of ML [Video 1-22]

Feature Engineering
What Feature Engineering Really Is
 What is a feature?
 Raw data vs features
 Feature engineering vs feature selection
 Why models don’t “understand” raw data
 Feature engineering as signal extraction
 When feature engineering matters more than algorithms

Understanding Data Before Engineering

 Feature types:
 Numerical
 Categorical
 Ordinal
 Text (high-level)
 Date & time
 Continuous vs discrete variables
 Cardinality (low vs high)
 Feature distributions
 Target leakage awareness

Missing Values Handling

 Why missing values occur
 Missing completely at random vs not random (intuition)
 Deletion strategies (row/column)
 Simple imputation:
 Mean / median / mode
 Constant value imputation
 Flagging missingness as a feature
 When not to impute
 Encoding Categorical Variables
 Why models can’t handle categories directly
 Label encoding (when it works, when it breaks)
 One-hot encoding
 Dummy variable trap
 Ordinal encoding
 High-cardinality categorical features (conceptual intro)
 Encoding leakage risks

Feature Scaling & Normalization

 Why scale features?
 Feature magnitude vs importance
 Standardization (z-score)
 Min–max scaling
 Robust scaling (intuition)
 When scaling is unnecessary
 Scaling before vs after train-test split (very important)

Handling Outliers
 What is an outlier?
 Outliers vs rare but valid values
 Visual detection (boxplots, scatter plots)
 Statistical detection (intuition only)
 Capping / flooring
 Removal vs transformation
 When outliers help the model

Feature Transformation
 Why transform features?
 Log transformation
 Power transformations (intuition)
 Handling skewed distributions
 Non-linear transformations
 Stabilizing variance
 Interpretability tradeoffs

Feature Creation (High-Impact Section)

 Domain-driven feature creation
 Ratios and differences
 Aggregated features (mean, count, sum)
 Interaction features (feature × feature)
 Binning / discretization
 Polynomial features (intuition)
 Temporal feature creation (high-level)
 Date & Time Feature Engineering
 Extracting year, month, day, weekday
 Cyclical features (sin/cos intuition)
 Time since / time until features
 Lag features (conceptual)
 Rolling window features (intuition)
 Avoiding time-based leakage

Text Feature Engineering (Intro Level)

 Why text needs special handling
 Text cleaning basics
 Tokenization (conceptual)
 Bag of Words intuition
 TF-IDF intuition
 When feature engineering gives way to embeddings
 What is deferred to NLP section

Feature Selection (Not Algorithms)

 Why fewer features can be better
 Redundant vs informative features
 Correlation-based filtering
 Variance thresholding (intuition)
 Domain-driven selection
 Selection vs dimensionality reduction (conceptual)
 Feature Engineering Pipelines
 Why pipelines matter
 Train-test consistency
 Fit vs transform distinction
 Order of operations
 Reproducibility
 Avoiding leakage via pipelines

Feature Engineering Pitfalls (Must-Know)

 Data leakage through features
 Encoding using full dataset
 Scaling using test data
 Over-engineering features
 Feature explosion
 Ignoring domain knowledge
Free Resources
 video100 Days of ML [Videos 23-45]

pending

ML Algorithms
Linear Models
 Linear Regression
 Multiple Linear Regression
 Polynomial Regression
 Ridge Regression
 Lasso Regression
 Elastic Net

Classification Algorithms
 Logistic Regression
 k-Nearest Neighbors (kNN)
 Naive Bayes
 Gaussian Naive Bayes
 Multinomial Naive Bayes
 Bernoulli Naive Bayes

Tree-Based Models
 Decision Tree (Classification)
 Decision Tree (Regression)

Ensemble Algorithms
 Random Forest
 Extra Trees (Extremely Randomized Trees)
 Boosting Algorithms
 AdaBoost
 Gradient Boosting
 XGBoost
 LightGBM
 CatBoost

Support Vector Machines

 Support Vector Classifier (SVC)
 Support Vector Regressor (SVR)

Unsupervised Learning Algorithms

 K-Means
 Hierarchical Clustering
 DBSCAN

Dimensionality Reduction
 Principal Component Analysis (PCA)
 Kernel PCA
 Linear Discriminant Analysis (LDA)
 T-sne
 Anomaly Detection
 Isolation Forest
 One-Class SVM
 Local Outlier Factor (LOF)

Probabilistic Models
 Gaussian Mixture Models (GMM)

Recommendation & Ranking

 Matrix Factorization
 Singular Value Decomposition (SVD)
Free Resources
 video100 Days of ML [Videos 45-130]

pending

ML Advanced Concepts
Hyperparameter Optimization
 Grid search
 Random search
 Bayesian optimization
 Model Evaluation & Validation
 Cross-validation strategies
 Stratified sampling
 Time-series validation
 Nested cross-validation

Imbalanced Data Handling

 Class imbalance problem
 Oversampling
 Undersampling
 SMOTE
 Class weighting
 Threshold moving

Ensemble Theory (Beyond Algorithms)

 Bagging vs boosting
 Stacking
 Blending
 Voting classifiers
 Interpretability & Explainability
 Model interpretability vs explainability
 Global vs local explanations
 Feature importance
 Partial dependence plots
 SHAP
 LIME

Model Selection & Comparison

 Bias vs variance driven selection
 Metric-driven model choice
 Statistical significance testing

Scaling Machine Learning

 Training on large datasets
 Incremental / online learning
 Distributed training (conceptual)
 Memory vs compute tradeoffs
Free Resources
 video100 Days of ML [Last 3-4 videos]

pending

End to End Projects

 Recommendation Systems
 Movie Recommendation System (Netflix-style)
 E-commerce Product Recommendation (Amazon-style)
 Online Course Recommendation
 News / Content Feed Ranking
 Music Playlist Recommendation

Prediction & Forecasting

 House Price Prediction
 Stock Price Movement Prediction
 Sales Forecasting System
 Demand Forecasting for Inventory
 Energy Consumption Forecasting

Classification Systems
 Spam Email Detection
 Credit Card Fraud Detection
 Loan Default Prediction
 Customer Churn Prediction
 Fake News Detection

Computer Vision (Classical ML / Intro DL)

 Handwritten Digit Recognition
 Product Image Classification
 Traffic Sign Recognition

NLP-Based Systems
 Sentiment Analysis System
 Resume Screening System
 News Article Categorization
 Text Similarity / Duplicate Detection

Anomaly & Risk Detection

 Fraud Detection System
 Network Intrusion Detection
 Insurance Claim Anomaly Detection
 Manufacturing Fault Detection

Customer & Business Intelligence

 Customer Segmentation System
 Market Basket Analysis
 Customer Lifetime Value Prediction
 User Behavior Analytics
 Healthcare & Social Impact
 Disease Risk Prediction
 Medical Cost Prediction

Famous “Interview Classic” Projects

 House Price Prediction
 Titanic Survival Prediction
 Customer Churn Prediction
 Credit Risk Modeling
 Recommendation System
Free Resources
 videoML Projects