Rust (2 blogmarks)

I like hearing about other people's AI coding workflows, especially if they are demonstrating concrete success on a complex task. Porting a browser engine from C++ to Rust in two weeks with no regressions is impressive. This is a great example of human heavily in the loop -- using knowledge of the system to pick what parts of the codebase to port and when.

I used Claude Code and Codex for the translation. This was human-directed, not autonomous code generation. I decided what to port, in what order, and what the Rust code should look like. It was hundreds of small prompts, steering the agents where things needed to go. After the initial translation, I ran multiple passes of adversarial review, asking different models to analyze the code for mistakes and bad patterns.

I thought this was going to be a sudoku thing at first, but it appears to be a problem with a bigger problem space that you have solve exhaustively because you need to find every solution in order to guarantee that you've found the largest GCD.

The code/solutions presented by the author are in Rust.