Thoughts and Writeups

by Davis Haupt

Software development is changing. Tool calling, inference scaling and RL with Verifiable Rewards have combined over the past year to enable agent harnesses like Claude Code which can reliably navigate, modify and contribute to large codebases.

LLMs scale amazingly well with the amount of training data you throw at them. But I’ve been thinking about how to build tools that work alongside the characteristics of LLMs rather than language models needing to learn how to work with existing human-centric tools during training.

I have a hunch that a programming environment built around the strengths and limitations of autoregressive LLMs can lead to cheaper and higher-quality agent-powered development. How could we prove out that hypothesis? One would first need to design a language that aligns with how LLMs “think”. What would such a language look like? In this post I put forward some ideas for a language called Markov that I think would fit the bill.

Read More...

2024 in review

Jan 4 2025

I wrote most of this post during the week between Christmas Eve and New Year’s Day when enough people in New York have left the city that it makes it quite difficult not to turn introspective. It’s been a pretty big year for me and for this blog, so I thought it would be fun to take a look back on the year as we all move into the second half of the decade.

Read More...

Welcome back to the Nix on Mac series! By the end of this post, you’ll be able to fully configure VSCode through your Nix flake via home-manager, which we set up in part 2. This includes:

  • Custom keybindings and settings
  • Installing themes and extensions from a nixpkgs overlay
  • Properly aliasing VSCode and other macOS applications to /Applications for Spotlight

Read More...

Dune: Part Two hit theaters recently, which got me thinking about my own experience reading Dune in the past, what I enjoyed about it, and my own personal headcannon that made this novel more enjoyable for me. And hey, it’s April Cools, so why not put pen to paper here?

Read More...