Posts - Perivitta Rajendran

Context Window Limits: Why Your LLM Still Hallucinates

Understanding token limits, retrieval failures, and why bigger context windows don’t automatically mean better answers

Posted by Perivitta on February 13, 2026 · 26 mins read

How to Generate Better Embeddings for Vector Search

A practical guide to improving retrieval quality with chunking, cleaning, metadata, and embedding strategies

Posted by Perivitta on February 12, 2026 · 21 mins read

Building Real-Time Chatbot Memory with Vector Databases + LLMs

How to give your chatbot long-term memory using embeddings, retrieval, and smart context management

Posted by Perivitta on February 12, 2026 · 27 mins read

Why Most RAG Systems Fail in Production

Common Pitfalls and Practical Fixes for Reliable Retrieval-Augmented Generation

Posted by Perivitta on February 10, 2026 · 17 mins read

A Beginner’s Guide to Building AI Safety Filters

How to Design and Implement Filters That Keep AI Content Safe and Responsible

Posted by Perivitta on February 10, 2026 · 12 mins read

← Newer Posts Older Posts →