SS / blog
Portfolio Tools
Writing

Blog

9 June 2026  ·  38 min read

Building Vectr, Part 1: Why grep Fails When You Don't Know the Keywords

How I built a local semantic code indexer for AI editors — covering AST chunking, hybrid vector+BM25 search, symbol graphs, and why naive approaches break on real codebases.

code-search semantic-search ast-parsing embeddings developer-tools mcp
Read →
24 May 2026  ·  40 min read

Why LLM Context Windows Fill Up Faster Than You Think — and What to Do About It

The token arithmetic behind system prompts, tool schemas, RAG retrieval, and conversation history — why the effective limit is below the advertised one, and four production strategies to manage it.

LLMs Context Windows Production AI Token Budget Deep Dive
Read →
24 May 2026  ·  35 min read

Why AI Code Assistants Waste Your Context Window — and How RAG Fixes It

A deep look at why attention dilution and positional bias make context stuffing counterproductive, and how AST chunking, hybrid BM25+dense retrieval, and RRF fusion fix it.

RAG Code Assistants LLMs Production AI Deep Dive
Read →
23 May 2026  ·  90 min read

The Complete Guide to Text Embeddings, Vector Databases & LLMs

From tokenization and transformers to cosine similarity, HNSW graph search, RAG pipelines, and LLM training costs. The deepest guide on the web — with 7 interactive demos.

Embeddings LLMs RAG Vector Search Deep Dive
Read →
22 May 2026  ·  30 min read

DPDP Act 2023 + AI/LLM: Automating Data Principal Requests in India

How AI and LLMs automate DPDP Act compliance — classifying data principal requests, drafting multilingual replies in 5 Indian languages, tracking 30-day SLAs, and generating audit-ready evidence trails.

DPDP AI / LLM India Privacy Compliance Open Source
Read →
© 2026 Swapnanil Saha
Part of the llm-tools suite