Make your dreams come true and consider hiring me!
You might also enjoy having an AI write your next compiler!
I recently discovered that I have some AI-supported things super power:I can implement now working Features – the ones I actually need and want – much faster than if I coded them artisanallyAnd there will be no loss in quality. I literally feel 10× more productive!
Take, for example, command history browsing, editing, and tab completion. ocamldebugBytecode debugger in the OCaml ecosystem. I’m talking about navigating through the command history with the arrow keys, going to the beginning of a line Ctrl-A and till the end Ctrl-Eand to get breakpoint information by typing
These are long standing features lldb And gdbprovided by libreadlineand not – are using rlwrap does No Provide the same experience.
Just last week, I wouldn’t have even dreamed of implementing such a feature. But a few days ago, I did exactly that. it was pr Completely AI-generated over the course of 2-3 days. Can you tell?
The code is organized into a series of small-sized commits. I challenge you to browse it and poke holes in the implementation!
This was not done in one fell swoop. I worked on it carefully – from evening till morning – because, honestly, it was funYes, I have enough OCaml experience (and I have over 30 years of development) to be sure that the code is basically what I would have written myself,
how i did it
i used cloud sonnet 4.5 to write code and chatgpt 5 To review it, I kept looping between them until I was satisfied. I started with the web versions of the models, then switched to the CLI to refine and polish the work – also because I’m on macOS, and the web models run on Linux. x86_64,
At one point, I was almost ready to roll up my sleeves and dive in manually. Cloud PTY got stuck on the issue and started making random changes which did not fix the problem. ChatGPT also didn’t help much.
Then I asked the cloud to show me the difference between the two code paths – yes, you can ask models to do this Describe the code! – And then success came.
Cloud, single-handedly, added debugging printouts to the problematic section of the code. Then it asked me to give it log output so it could troubleshoot more effectively. We repeated this several times until Claude finally found the root cause and fixed it.
However, one caveat: Even if I didn’t type the code myself, own This – and this is my responsibility now.
Feeling
This process takes time – although it’s still faster than coding by handIt feels a bit like being a lead engineer guiding a team of mid to junior level developers, I can coordinate multiple projects at once, but my real bottleneck is cognitive bandwidth: the need to review, understand, and reason about them all at once,
Overall, I feel like a coding god. There are so many things I can create now – so many problems I can tackle – that I wouldn’t have tried before. The sky is literally my coding limit.
I, in a way, welcome our new AI overlords. I don’t care who wrote the code – human or machine – as long as it’s clean and it works.
Do you care who created your favorite software features, or just that they shipped fast and worked great?
Let me know on Hacker News, Twitter or the OCaml forum!
PS – Use AI to learn
yes you can do And Needed Use AI to learn. I do.
As an autodidact, this is a perfect fit for me. You can ask AI to explain an unfamiliar codebase in whole or in part, or to show you variations, optimizations, or tests.
There’s nothing standing between you and your new learning superpowers!
PS – AI will write your next compiler!
I’m working on adding DWARF debugging information to the OCaml compiler. Just for kicks and because I don’t necessarily want to use OxCaml, the Gen St fork of OCaml that supports DWARF on Mac.
DWARF is the information you need to see the source code lldb Or gdbInspect variables, stop at lines of code, etc.
I seem to have got it working, but I want to make absolutely sure before posting about it.
<a href