Deep dives, learnings, & technical logs
I started with Figma and design systems. Two years later I'm writing FastAPI services and PostgreSQL migrations. Here's what the transition actually looked like — the good parts and the uncomfortable ones.
A practical guide to configuring pgvector on a self-hosted PostgreSQL database, structuring semantic database queries, and calculating high-performance cosine similarity vector distances.
A candid analysis of migrating a production portfolio to Next.js 16 (App Router) and React 19, exploring the benefits of server-side data fetching, Turbopack compiling speed, and practical strategies for solving hydration mismatches.
A complete deep-dive into building a production-ready, low-latency Retrieval-Augmented Generation chatbot from scratch using sentence-transformers, PostgreSQL asyncpg connection pooling, and Gemini.
Most RAG tutorials show the happy path. Here's what actually happens when you build one for production — the rate limits, the embedding cold starts, the hallucinations, and how I handled all of it.
Users were getting logged out randomly and API calls were returning 401s. The cause took 2 hours to find and 5 minutes to fix.
Shared hosting looked cheaper until the firewall started blocking outbound API calls. Here's the full migration story and everything that broke along the way.
The exact steps I took to untangle a broken multi-provider DNS setup that was causing intermittent 404s and SSL errors in production.