How to port a PHP project to Rust with Copilot
The author reports 40,000 generated Rust lines and estimates 90% progress after two weeks of nightly iteration, while explicitly expecting additional fixes before the result works.
news.ycombinator.com
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Port [existing PHP project] to Rust using [selected frameworks] and [design constraints]. Work through [manageable feature chunks], track missing functionality against the original, and report what remains untested or incomplete after each iteration.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- plqbfbv
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The author reports 40,000 generated Rust lines and estimates 90% progress after two weeks of nightly iteration, while explicitly expecting additional fixes before the result works.
- Human involvement
- The author directed design and framework choices and expected to fix problems manually or ask Copilot for further fixes.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Asked Copilot Agent mode to translate the existing PHP project into Rust.
- Broke the rewrite into manageable chunks.
- Directed design choices and selected specific frameworks.
Plqbfbv experimented with Copilot Agent mode on a web-based PHP project needed by their wife. The project had not been maintained for a while. The author asked Copilot to translate it into Rust, initially treating the exercise as an experiment and expecting little from it.
They broke the work into manageable chunks, directed design choices and requested specific frameworks. Copilot then wrote Rust code and tracked functionality still missing compared with the original project. The author reports watching it iterate independently for more than 30 minutes at a time.
After two weeks of spending one to two hours each night on the work, the author reported 40,000 generated Rust lines and estimated they had about 90% of the intended result. These are progress claims rather than evidence that the rewritten application was ready for use. They explicitly expected the first attempt not to work and anticipated either manual fixes or further direction to Copilot.
The author identifies their background as systems and cloud engineering rather than programming. They found syntax boilerplate and reduced documentation searching particularly valuable, while continuing to direct the rewrite's design and implementation choices.
Who it's for: technically experienced users exploring a supervised migration of an existing application into another language.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · plqbfbv
Frequently asked questions
Was the rewritten application confirmed working?
No. The author expected it would not work on the first attempt and anticipated manual fixes or further Copilot-directed corrections.
How did the author guide Copilot?
They divided the work into manageable chunks, directed design choices and asked for specific frameworks.
More details
Remaining steps
- Iterated nightly while Copilot generated code and tracked missing functionality.
About this account
Copilot generated Rust code, tracked gaps against the original project and sometimes iterated independently for over 30 minutes.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=44542053. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: plqbfbv on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
GitHub Copilot Agent mode · PHP · Rust
PHP · Rust · Application rewrite
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.