A structured roadmap from zero to production AI systems. Every concept has an interactive demo — try it first, then read the numbered explanation below.
The bedrock everything else depends on.
NumPy, Pandas, Matplotlib — the daily tools of every AI engineer.
Linear algebra gives you intuition for how models transform data. Calculus explains how they learn.
Version control, virtual environments, command-line fluency.
Learn how models learn.
Regression, classification, cross-validation, train/test splits, evaluation metrics.
Perceptrons, forward pass, backpropagation, activation functions, gradient descent.
The architecture that powers modern AI.
Convolutional filters, pooling, pre-trained models (ResNet, EfficientNet).
The architecture behind GPT, Claude, Gemini — every modern LLM.
From model weights to products people use.
LLMs don't read words — they read tokens. BPE, SentencePiece, context windows.
Temperature, top-k, top-p sampling — the controls that shape LLM output.
Why LLMs confidently generate falsehoods, and how to prevent it.
How words become vectors. Cosine similarity, semantic clusters, embedding models.
Ship AI that works in the real world.
Ground LLMs in real documents. Chunking, vector retrieval, prompt injection.
Observe → Plan → Act → Reflect loops. Tool use, multi-agent systems, memory.
LLM-as-judge, RAGAS, PromptFoo, regression suites. Ship only what you can measure.
FastAPI, Docker, monitoring, cost optimization, CI/CD for LLM systems.
Ship things. Get hired.
Employers care about shipped products, not certificates. 3 projects by difficulty.
Public GitHub + live demo + metrics in the README + blog post = a winning portfolio.