Featured article
Build a Reliable Python AI Agent with Structured Tool Calling
Schemas, timeouts and tests for a tool-calling agent you can run and verify.
Rajath Kumar K S
Engineering blog
Technical guides from working code
Tested, version-pinned tutorials on AI Agents, MCP, RAG and Python backends. Every guide links to runnable code and states its limits.
Articles
Browse by topic
MCPMCP with Python: tools, resources and permission boundaries
How to expose tools and resources through the Model Context Protocol, and where to draw the permission line.
12 min readIntermediate
Rajath Kumar K S
RAGProduction RAG: chunking, retrieval and citations that hold up
The three decisions that decide whether a retrieval-augmented answer can be trusted.
15 min readIntermediate
Rajath Kumar K S
Evaluation & TestingEvaluating LLM outputs with a reproducible test matrix
Turn vague 'looks good' checks into a small test matrix you can rerun after every change.
12 min readIntermediate
Rajath Kumar K S
FastAPI & BackendServing an AI endpoint with FastAPI and async Python
A small, safe FastAPI service in front of a model call: validation, timeouts and limits.
11 min readIntermediate
Rajath Kumar K S
SecurityPrompt injection and tool permissions: a practical defence
Why untrusted text can steer your agent, and the permission rules that limit the damage.
13 min readIntermediate
Rajath Kumar K S