# Building a C compiler with a team of parallel Claudes

**URL:** <https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698>\
**Category:** Uncategorized\
**Created:** [February 6, 2026, 10:13am UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698 "2026-02-06T10:13:29Z")\
**Posts on this page:** 16\
**Page:** 1

<div class="post-metadata">

**Author:** ![zaikunzhang](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/zaikunzhang/32/785_2.png) [@zaikunzhang](https://fortran-lang.discourse.group/u/zaikunzhang)\
**Post date:** [February 6, 2026, 10:13am UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/1 "2026-02-06T10:13:29Z")

</div>

> **[Building a C compiler with a team of parallel Claudes](https://www.anthropic.com/engineering/building-c-compiler)**
>
> Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

---

<div class="post-metadata">

**Author:** ![jorgeg](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/jorgeg/32/6835_2.png) [@jorgeg](https://fortran-lang.discourse.group/u/jorgeg)\
**Post date:** [February 6, 2026, 11:21am UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/2 "2026-02-06T11:21:16Z")

</div>

Maybe if we all got together and spent 100k USD we can build slopfort the AI built Fortran compiler

This is an insane demonstration of what unsupervised agentic coding can do. Although insane, there are limits to these things.

---

<div class="post-metadata">

**Author:** ![R\_cubed](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/r_cubed/32/263_2.png) [@R\_cubed](https://fortran-lang.discourse.group/u/R_cubed)\
**Post date:** [February 6, 2026, 3:22pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/3 "2026-02-06T15:22:09Z")

</div>

All of this is “interesting” until you factor in that:

1. the human brain can do all of this (and more!) with much less energy.

[https://medium.com/write-a-catalyst/human-brains-beat-ai-by-225-000-times-in-energy-efficiency-762b9327e8ad](https://medium.com/write-a-catalyst/human-brains-beat-ai-by-225-000-times-in-energy-efficiency-762b9327e8ad)

1. Far too often, the most hyped results in the media are entirely fake.

[![](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/5/5fbf06ebe4fe46c686a4481a4888988674374193.jpeg "Debunking Devin: "First AI Software Engineer" Upwork lie exposed!") ](https://www.youtube.com/watch?v=tNmgmwEtoWE)

These substitutes for human thinking and skill are directed by people who fundamentally **dislike** programming, and \*\*dislike\*\* dealing with human beings who happen to have thoughts, wishes, and goals of their own.

---

<div class="post-metadata">

**Author:** ![Beliavsky](https://avatars.discourse-cdn.com/v4/letter/b/ba8739/32.png) [@Beliavsky](https://fortran-lang.discourse.group/u/Beliavsky)\
**Post date:** [February 6, 2026, 3:23pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/4 "2026-02-06T15:23:41Z")

</div>

My list of [Fortran tools](https://github.com/Beliavsky/Fortran-Tools) has a few projects that translate Matlab to Fortran:

[matlab2fmex](https://sourceforge.net/projects/matlab2fmex/): small translator which aims to convert numerical Matlab m-files to Fortran90 mex files, by Ben Barrowes. matlab2fmex first translates an m-file to a Fortran90 mex source file then compiles that Fortran90 file using Matlab’s mex and the local compiler.

[matlab2fortran](https://github.com/ebranlard/matlab2fortran): performs some simple conversions from Matlab code to Fortran, by ebranlard

[Mc2For](https://github.com/Sable/Mc2For): MATLAB to Fortran compiler, from Sable

Building on one or more of them with Claude Code or OpenAI Codex, it may be possible to translate a large fraction of Matlab code online to Fortran. There is [GitHub - pyccel/pyccel: Python extension language using accelerators](https://github.com/pyccel/pyccel) to translate Python with NumPy to Fortran and [GitHub - t-kalinowski/quickr: R to Fortran Transpiler](https://github.com/t-kalinowski/quickr) to translate R to Fortran. There should be tools so that all numerical code in Matlab, Python, and R is translatable to Fortran.

---

<div class="post-metadata">

**Author:** ![tyranids](https://avatars.discourse-cdn.com/v4/letter/t/3e96dc/32.png) [@tyranids](https://fortran-lang.discourse.group/u/tyranids)\
**Post date:** [February 6, 2026, 3:58pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/5 "2026-02-06T15:58:05Z")

</div>

I dunno, this sounds pretty amazing to me. You have to remember that right now is the worst AI is going to be. It only gets better from here. December 2022 was the first time the AI chatbot could kind of make a haiku, and now people are working parallel teams of them to do months worth of human work in 2 weeks at a fraction of the cost.

---

<div class="post-metadata">

**Author:** ![kimala](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/kimala/32/2707_2.png) [@kimala](https://fortran-lang.discourse.group/u/kimala)\
**Post date:** [February 6, 2026, 4:29pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/6 "2026-02-06T16:29:12Z")

</div>

> [@tyranids](#):
>
> at a fraction of the cost

I would like to see the cost comparison though

---

<div class="post-metadata">

**Author:** ![R\_cubed](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/r_cubed/32/263_2.png) [@R\_cubed](https://fortran-lang.discourse.group/u/R_cubed)\
**Post date:** [February 6, 2026, 4:55pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/7 "2026-02-06T16:55:32Z")

</div>

> [@tyranids](#):
>
> I dunno, this sounds pretty amazing to me. You have to remember that right now is the worst AI is going to be. It only gets better from here.

That is the biggest lie in AI, as Carl Brown discusses in this video (less than 10 min long).

[![](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/c/c1cd3217c773ec5d8ab8a6a3ae2887aae896c2d9.jpeg "The Biggest Lie in AI") ](https://www.youtube.com/watch?v=0Plo-zT8W9w)

---

<div class="post-metadata">

**Author:** ![tyranids](https://avatars.discourse-cdn.com/v4/letter/t/3e96dc/32.png) [@tyranids](https://fortran-lang.discourse.group/u/tyranids)\
**Post date:** [February 6, 2026, 5:17pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/8 "2026-02-06T17:17:45Z")

</div>

It was $20k

> Over nearly 2,000 Claude Code sessions across two weeks, Opus 4.6 consumed 2 billion input tokens and generated 140 million output tokens, a total cost just under $20,000.

Perhaps it can be done in other jurisdictions, but no one in America is going to build a compiler tabula rasa in 2 weeks for $20,000 USD with human software engineers.

---

<div class="post-metadata">

**Author:** ![wspector](https://avatars.discourse-cdn.com/v4/letter/w/47e85d/32.png) [@wspector](https://fortran-lang.discourse.group/u/wspector)\
**Post date:** [February 6, 2026, 6:12pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/9 "2026-02-06T18:12:13Z")

</div>

Saw this on X today:

 ![image](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/2/2dccec6afb748231177ad2142fc14ab841da3c82.jpeg)

---

<div class="post-metadata">

**Author:** ![wspector](https://avatars.discourse-cdn.com/v4/letter/w/47e85d/32.png) [@wspector](https://fortran-lang.discourse.group/u/wspector)\
**Post date:** [February 6, 2026, 6:16pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/10 "2026-02-06T18:16:01Z")

</div>

Eric Raymonds take:

 ![image](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/a/a14542d0b5e89c0f4ed4f06108c0255fa136f6dc.png)

---

<div class="post-metadata">

**Author:** ![wspector](https://avatars.discourse-cdn.com/v4/letter/w/47e85d/32.png) [@wspector](https://fortran-lang.discourse.group/u/wspector)\
**Post date:** [February 6, 2026, 6:20pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/11 "2026-02-06T18:20:52Z")

</div>

But then Eric also writes:

 ![image](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/f/f8290f3dc6c255c25265fbef5110b8bc22cffbe2.png)  
 ![image](https://global.discourse-cdn.com/free1/uploads/fortran_lang/original/2X/7/7dd74e701aedcaedab5d1f956d6c4b9b840481ab.png)

---

<div class="post-metadata">

**Author:** ![certik](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/certik/32/4_2.png) [@certik](https://fortran-lang.discourse.group/u/certik)\
**Post date:** [February 6, 2026, 11:04pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/12 "2026-02-06T23:04:29Z")

</div>

If you want to help Fortran, take LFortran and run it on your code, it will fail somewhere, then ask Codex or Claude Code to create an MRE (minimal reproducible example), then open up an issue at LFortran. We fix MREs very quickly. The bottleneck right now is to create them. AI is actually super helpful for that.

AI is also super helpful in fixing MREs — but please stay in the loop, don’t submit a PR without understanding what it does.

As to the future, there is a lower bound and an upper bound.

The lower bound is that the underlying technology will not fundamentally improve, so it will be like a steam engine: the engine didn’t exponentially improve, but the harness around it did: the first machines were really slow, but the steam engines from 1940s were really fast (100 mph). So LLMs will be able to run locally (since the hardware almost for sure will improve) and the agents will improve etc. I can’t currently afford it, but would love to have an agent running 24/7, just triaging bugs and fixing what it can fix and submitting PRs, and have a large team of agents being able to do what I ask them. The technology is already there, so this will almost for sure happen. An experienced developer will have to be in the loop to deliver.

The upper bound is that the technology will improve exponentially. Then of course inevitably LLMs will be better than humans at coding.

---

<div class="post-metadata">

**Author:** ![RonShepard](https://avatars.discourse-cdn.com/v4/letter/r/a3d4f5/32.png) [@RonShepard](https://fortran-lang.discourse.group/u/RonShepard)\
**Post date:** [February 6, 2026, 11:41pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/13 "2026-02-06T23:41:23Z")

</div>

> [@certik](#):
>
> Then of course inevitably LLMs will be better than humans at coding.

Somewhere along the line LLMs will start coding themselves, and then Skynet will happen. We’ve already seen the movie!

---

<div class="post-metadata">

**Author:** ![ZaneHam](https://avatars.discourse-cdn.com/v4/letter/z/51bf81/32.png) [@ZaneHam](https://fortran-lang.discourse.group/u/ZaneHam)\
**Post date:** [February 7, 2026, 3:33am UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/14 "2026-02-07T03:33:41Z")

</div>

‘slopfort’, yet another word to chuck into the lexicon.

---

<div class="post-metadata">

**Author:** ![jwmwalrus](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/jwmwalrus/32/4483_2.png) [@jwmwalrus](https://fortran-lang.discourse.group/u/jwmwalrus)\
**Post date:** [February 7, 2026, 4:52am UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/15 "2026-02-07T04:52:20Z")

</div>

I remember one episode of the TV series Numb3rs, in which an AI sponsored by DARPA seemed to have committed murder. The mathematicians eventually find out that all the “AI” does is passing the Turing test over and over (and it was the victim’s wife who actually committed the murder).

LLMs are impressive, but in essence they’re just that (i.e., Turing test interfaces). They get some things right and some things wrong, but that’s only a probabilistic consequence based upon the data they were trained on.

The narrative of the companies invested in the AI bubble is that by just adding more computing power the AI singularity will eventually be achieved —which is akin to saying that by adding more pages to a dictionary, it will eventually become sentient.

---

<div class="post-metadata">

**Author:** ![R\_cubed](https://yyz2.discourse-cdn.com/free1/user_avatar/fortran-lang.discourse.group/r_cubed/32/263_2.png) [@R\_cubed](https://fortran-lang.discourse.group/u/R_cubed)\
**Post date:** [February 7, 2026, 4:12pm UTC](https://fortran-lang.discourse.group/t/building-a-c-compiler-with-a-team-of-parallel-claudes/10698/16 "2026-02-07T16:12:24Z")

</div>

A few observations:

1. For all of the work spent developing a non-executable context document, he could have written a formal spec, and then derived an implementation, providing a much higher level of assurance that the code is correct by construction (with regard to the specification) and a plan to test both the spec and implementation. Much of that could be automated using a high level language like Prolog, ML, or Haskell, then generating imperative code (C, C++, etc) from the prototype (executable spec and reference implementation). I don’t see how any of this wrestling with LLMs using a context doc is an economic value add.
2. Without a spec, what does it mean to say “The LLM produced correct code”? All that post gives is vague, subjective impressions. When any of this is put to a test, the use of LLM for coding appears to slow people down, vs. the expectation of speeding things up.

> **[Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer...](https://arxiv.org/abs/2507.09089)**
>
> Despite widespread adoption, the impact of AI tools on software development in the wild remains understudied. We conduct a randomized controlled trial (RCT) to understand how AI tools at the February-June 2025 frontier affect the productivity of...

> When AI tools are allowed, developers primarily use Cursor Pro, a popular code editor, and Claude 3.5/3.7 Sonnet. Before starting tasks, developers forecast that allowing AI will reduce completion time by 24%. After completing the study, developers estimate that allowing AI reduced completion time by 20%. **Surprisingly, we find that allowing AI actually increases completion time by 19%–AI tooling slowed developers down. This slowdown also contradicts predictions from experts in economics (39% shorter) and ML (38% shorter).**

Take their results for whatever you think they are worth, but at least they tried to measure something tangible.
