Blog
Articles on applied AI, product engineering, search, and delivery. Writing about RAG systems, agentic AI, and building products that scale.
- Deploying GLM-OCR as a Serverless GPU API on Modal (for almost $0)
A quick run through on how to get a leading OCR model deployed in a server-less environment
- · How to build a minimal viable coding agent
A quick exploration into building a coding agent in around 100 lines of code
- · A guide to model quantization in fine-tuning (and how to pick the right GGUF)
Stop guessing which GGUF to use. A robust mental model for model quantization, helping you balance VRAM, perplexity, and performance when exporting fine-tuned models for local inference.
- · Fine-tuning an open source LLM to speak Doric
Practical case study on fine-tuning a small open-weight LLM with LoRA to speak a nice Scottish dialect, from dataset creation and Unsloth training through evaluation to deployment with vLLM/Modal.
- My Python starter-kit
A modern Python starter kit with TypeScript-like developer experience I use for my experiments.
- AI Tools I use: Nuanced
Nuanced is a local code analysis tool that gives engineers and LLMs an understanding of how code actually behaves
- Introducing Fetch Engines: reliable markup and content fetching
Fetch Engines is an open source extraction toolkit that turns web pages into high-quality Markdown and structured JSON data.
- How I built this site
A walkthrough of the architecture, content pipeline, styling, SEO, and tooling behind this personal site and blog.
- Claudette Patterns for TypeScript: A Guide to the AI SDK
A guide to recreating Claudette's ergonomic features (like tool loops and structured output) in TypeScript using the Vercel AI SDK.
- Hello World
This is where I write about my learning, experiements, and other things.