Fred McDavid
Hi, I’m Fred McDavid. I’m a backend software engineer, maker, and taker-on of Projects. Professionally, I design fast systems that are stable and well-tested. Lately, I use AI agents to build such things much more quickly than I used to.
Outside of work, I build fences, rehab old farm structures back into usable shape, and keep the trails on our farm clear enough for our boarders to ride on. When I can, I make time to catch a nearby concert or brew a batch of beer.
My wife and I run Hoppin’ Fox Equestrian, a small horse-boarding business on our farm, and we operate a handful of rental properties through Landmetrics, our residential real estate business.
This site is home base, mostly links to elsewhere.
Framework Labs Posts
Framework Labs is where I post on AI and related topics
Language Is All You Need
A lot of the sessions I sat in at beCamp this month were wary of AI assistants. It was Charlottesville’s tech unconference, celebrating its twentieth anniversary at UVA’s School of Data Science, and …
Julia 1: A Model That Only Answers Multiple Choice
I recently came across Julia 1, a “decision model” from Supersonic Labs on the Google newsfeed. It does the same kind of job as TypeSafe’s Jev, and it isn’t an LLM at all. Give it some context, a …
The Filesystem Your Agent Already Speaks
A lot of the ideas in this piece on the new Ambiance harness seem obvious as I read the post, but not in a bad way. It’s more like giving voice to a few things that have been bouncing around in my …
ECC: 228K Stars for an Agent Harness
I read a piece on the ECC agent harness and I did a bit of a double-take on the GitHub stars: 228,000 stars for a Claude Code configuration system built by one person over ten months. Not a new model, …
GraphRAG and the Pyramid I'd Only Half-Built
Retrieval-augmented generation (RAG) has a bit of a problem with the relations between the chunks. I’ve thought about a few solutions and the one I think is the most promising involves a series of …