It does not matter what 80-90% of developers do. Code development is heavily tail-skewed: focus on the frontier and on the people who are able to output production-level code at a much higher pace than the rest.
What’s the solution here, reward code that works without try catch, reward code that errors and is caught, but penalize code that has try catch and never throws an error?
The gap between coding agents in your terminal and computer agents that work on your entire operating system is just too narrow and will be crossed over quick.
> Originally, it was all about quality data sources.
It still is! Lots of vertical productivity data that would be expensive to acquire manually via humans will be captured by building vertical AI products. Think lawyers, doctors, engineers.
The moat is people, data, and compute in that order.
It’s not just compute. That has mostly plateaued. What matters now is quality of data and what type of experiments to run, which environments to build.
Stacked diffs is a huge one, and also where improving git would also improve LLM workflows. The bottleneck after code generation is PR reviews, and stacked diffs help break down large PRs into more digest-able pieces.
If you help humans collaborate better, you help LLMs collaborate better.
Kids in college are also lonely; I go to a school with a relatively large undergraduate populations. The result is that many people have trouble finding friends given that interactions are less frequent. Dorms and free time do help, but in the first week there were a lot of posts asking others on how they made their friends.
also works if you have the GitHub cli installed. Would setup an AGENTS.md or SKILL.md to instruct an agent on how to use gh too.