A structured, hands-on study guide covering everything from Python basics to deploying production ML models — designed for all skill levels.
No prior coding experience? Start from zero with Python fundamentals and build up gradually.
Supplement your university coursework with practical, applied AI/ML examples and projects.
Upskill and transition into data science or AI roles with job-ready projects and knowledge.
Apply modern ML/DL frameworks to your research domain with deeper dives into theory.
A clear progression from first principles to advanced systems — follow in order or jump to what you need.
Build the essential skills every ML practitioner needs. No shortcuts — these will carry you for the entire journey.
Learn the full machine learning pipeline — from data prep to model evaluation — using Scikit-learn and real datasets.
Build neural networks, transformers, and LLM-based systems. Deploy everything to production with modern MLOps tooling.
Each module is carefully structured with theory, code examples, exercises, and a capstone project.
Variables, data types, control flow, functions, OOP, comprehensions, and file I/O — all through ML-relevant examples.
Linear algebra, calculus intuition, probability, statistics, and distributions — using NumPy and SciPy with visual explanations.
Load, clean, transform, and merge real-world datasets. Handle missing values, outliers, and feature engineering at scale.
Tell stories with data using Matplotlib, Seaborn, and Plotly. EDA techniques, distribution plots, and interactive charts.
Supervised and unsupervised algorithms, model evaluation, hyperparameter tuning, and production-ready pipelines.
Build neural networks from scratch. CNNs for vision, RNNs for sequences, training loops, GPU acceleration, and transfer learning.
Text preprocessing, TF-IDF, word embeddings, fine-tuning BERT, building with HuggingFace Transformers, and working with LLMs.
Serve models with FastAPI, containerize with Docker, CI/CD for ML, monitoring, versioning with MLflow, and cloud deployment.
Apply everything you've learned across 5 end-to-end projects: from a movie recommender to a production-grade RAG chatbot.
Every project uses real datasets and mirrors what engineers actually build in industry.
Regression model on the Boston Housing dataset. Feature engineering, cross-validation, and a simple web UI.
NLP pipeline: tokenization → TF-IDF → Naive Bayes classifier. Achieves 97%+ accuracy on the Enron dataset.
Collaborative filtering + content-based hybrid recommender using the MovieLens dataset and cosine similarity.
Train a CNN on CIFAR-10 from scratch, then fine-tune ResNet-50 with transfer learning for custom image classification.
Time-series forecasting using LSTM networks on historical stock data. Includes feature engineering and evaluation.
Build a document Q&A system using retrieval-augmented generation with LangChain, FAISS, and a local Ollama LLM.
Download or bookmark these one-page references — designed for quick lookups while you code.
Core syntax, data types, loops, functions
Arrays, broadcasting, linear algebra ops
DataFrames, groupby, merge, reshape
Plots, subplots, styling, saving figures
Models, pipelines, metrics, grid search
Tensors, autograd, nn.Module, training loop
Transformers, tokenizers, fine-tuning
Gradient descent, loss functions, metrics
Carefully selected books, courses, datasets, and tools — no fluff, only what actually helps.
Everything you need to go from zero to job-ready — structured, free, and always up to date.