Week 1 kicks off with a foundations refresher covering AI/ML/LLM basics, Python & API setup, and the core model-retrieval-tool-memory stack. You'll build your first AI app

Learn system prompt design, strict output schemas, end-to-end function/tool calling, choosing the right model/provider, and classification & routing logic. Build a bot that classifies incoming requests and routes them automatically.

Learn embeddings & cosine similarity, chunking strategies, vector databases & indexing, semantic retrieval, and building an ingestion pipeline. Build a semantic search tool over a set of documents.

Learn the full RAG pipeline end to end, grounding answers in retrieved sources, generating trustworthy citations, measuring groundedness, and fallback behavior. Build a Q&A assistant that cites its sources.

Learn query transformation (HyDE), context engineering, building a golden test dataset, before/after benchmarking, and turning results into a report. Upgrade your Citation Assistant and prove the improvement with data.

Learn workflows vs agents, the agent loop built from scratch, ReAct/reflection/planning patterns, agent state management, and stop conditions. Build an autonomous agent that completes multi-step tasks.

Learn prompt injection attacks & defense, PII detection & redaction, a responsible-AI checklist, containerizing with Docker, and deploying live. Harden and deploy your assistant to a public URL.

Bring together everything from Weeks 1–7 — streaming, structured output, RAG, agents, and guardrails — into one system built around a chosen domain and track. Present it live to a panel, get scored, and receive feedback plus your certificate.