Personal RAG

2026

A RAG system to make an AI answer questions about my professional career at the chat feature on my website.

RAGBackendTypeScriptAILLM

Overview

A custom Retrieval-Augmented Generation (RAG) backend developed to power the interactive AI chat on this portfolio, serving precise information about my career, skills, and projects.

The Challenge / Problem

I wanted to make the LLM answer questions about my professional career and projects more precisely, dropping down hallucinations that could be caused by simply telling the AI what to say through prompts.

The Solution

Built a specialized RAG pipeline using vector embeddings and semantic search to retrieve relevant career context from verified portfolio documents before feeding it to the LLM.

Project Highlights

Vector embeddings and semantic retrieval for precise career Q&A
Custom chunking and indexing of professional experience and case studies
Fast low-latency API route integrated into the portfolio AI chat modal