Ludmal De Silva — AI Engineer
Hi, I'm Ludmal. I write code for a living and write about code for fun — a habit my family has politely flagged as “a bit much.”
For the last decade I've been living in the AI side of things, mostly RAG and AI-assisted code generation. I've authored a small library of internal skills that help dev teams get more out of their AI tools than yet another console.log autocomplete. Some of them have shipped. One of them I'm proud of. The rest I'm politely not talking about.
Agents That Pay for Themselves: The Economic Loop That Will Redefine Software in 2027
Most agents lose money on every call. A handful have started covering their own inference cost by generating measurable value per run. The structural difference is worth understanding before you build the next one.
From RAG to Production — Observability, Cost Controls, and the Reality No Demo Shows
Everything the tutorials skip. The instrumentation, the kill switches, and the 3am pager habits that turn a RAG demo into something you can keep on call.
The Vibe Coder's Dilemma: When to Read the Code You Just Generated
Reading every AI-generated line is slow. Reading none is reckless. The honest answer is "it depends," and the dependency is more predictable than people admit.
ludmal/microsoft-agent-framework-sample
Microsoft Agent Framework — Content Pipeline Sample. A tiny ASP.NET Core Web API that wires up three agents with the Microsoft Agent Framework and runs them as a sequential workflow. Send a topic, the agents pass their work down the line, and you get a polished article back.
ludmal/rag-pinecore-sample
RAG + Pinecone — a working Next.js tutorial. A small but realistic Next.js 15 app that walks through every step of a RAG pipeline: upload documents, watch them get chunked and embedded, store the vectors in Pinecone, and chat with Claude using those documents as grounding.
ludmal/pylib
Python library for useful tools. Includes utilities like email templates and an SMTP mail-sending helper.