Week 1 – Python for Agents

Idiomatic Python, OOP, decorators, API & CLI tools, JSON handling.
Lab: Data extractor + API client

Week 2 – NLP for LLMs

Tokenization, NER, POS, Word2Vec, Transformers, parsing.
Lab: NLP pipeline (spaCy)

Week 3 – Embeddings & Search

TF-IDF → Transformers, FAISS, Chroma, vector search.
Lab: Build semantic search engine

Week 4 – LLM Internals

Attention, Transformer models, open vs closed LLMs.
Lab: Use Hugging Face & OpenAI APIs

Week 5 – Using LLM APIs

OpenAI, Claude, Mistral, Groq, chat vs completion, streaming.
Lab: Integrate GPT/Claude into web app

Week 6 – Prompt Engineering

Prompt types, COT, ReAct, JSON outputs, safety patterns.
Lab: Prompt wrapper for complex tasks

Week 7 – Retrieval-Augmented Generation (RAG)

Architecture, query rewriting, hybrid search, fallback logic.
Lab: RAG app with docs + LLM

Week 8 – Memory & Context

Short vs long-term memory, summarization, vector memory.
Lab: Context-aware chat agent

Week 9 – Architecting Agents

Reactive vs planning agents, memory, retries, fallback chains.
Lab: Task planner agent (LangChain)

Week 10 – Autonomous Agents

Goal decomposition, BabyAGI, AutoGPT, planning loops.
Lab: Research agent with auto-execution

Week 11 – Reflection Agents

Tree of Thought (ToT), self-debugging, self-evaluation loops.
Lab: Reflective + replanning agent

Week 12 – Multi-Agent Systems

Planner-Worker, Debate agents, shared memory coordination.
Lab: 3-agent collaborative system

Week 13 – Model Context Protocol (MCP)

MCP schema, structured context, tool results & memory sharing.
Lab: Build MCP-compliant agent tool

Week 14 – Deployment & APIs

Docker, serverless, cloud functions, exposing agents via APIs.
Lab: Deploy agent via Flask/n8n

Weeks 15–16 – Safety, Ethics & Final Project

Monitoring (LangSmith), guardrails, rate limits, evaluations.
Lab: Add safety + observability to agent project