🐍 Python-For-AI
Python AI & Agentic Systems track — part of my 8-month transition into
AI Infrastructure Engineering.
While my HKB-goes-Cloud handles cloud-native infrastructure, this
repo is where the intelligence lives: agent orchestration, RAG pipelines,
LLM evaluation, fine-tuning, and MCP server development.
What's Inside
Months 1–2 — Python fundamentals, async, FastAPI, OpenAI SDK,
LangChain basics, ChromaDB vector store
Months 3–4 — LangGraph stateful agents, CrewAI multi-agent crews,
AutoGen GroupChat, Pinecone/Weaviate, RAG with HyDE + MMR retrieval
Months 5–6 — RAGAS + LangSmith + TruLens evaluation pipelines,
semantic caching, AI cost routing (mini/full/local model selection),
LoRA/QLoRA fine-tuning on biotech domain data
Months 7–8 — MCP server exposing biotech tools via universal
protocol, OpenAI Agents SDK, A2A protocol awareness, multi-agent
platform with 5+ coordinating agents, production observability +
agent debugging
Stack
Python 3.12 · FastAPI · LangChain · LangGraph · CrewAI ·
AutoGen · OpenAI Agents SDK · MCP SDK · ChromaDB · Pinecone ·
Weaviate · RAGAS · LangSmith · Hugging Face · PyTorch ·
Ollama · vLLM · Pydantic · Instructor
Why Two Repos
Go for systems. Python for AI. Same architecture used at Datadog,
Anthropic, and Google for production AI infrastructure. This isn't a
choice — it's how the industry builds.
_Part of my public learning journey →
· Go Repo · X/Twitter
