Viewing profile — bthornbury

lots of good ones still

I discuss the testing approach and coverage with the model before and after, sometimes in a fresh thread that does a static analysis. interestingly my input is still pretty importa…

codex pro plan currently

for me, almost all of the work is specs I am no longer: - reading docs for hours and hours - typing (barely at all) - writing code - manually doing tight debug loops - using an IDE…

we need some better standard long-context benchmarks. needle in a haystack is not good for this, yes it proves the model can attend to its context, but in its usual form, somewhat …

the qwopus 27b model is good for grunt work style tasks, even across multiple files. Piping a bunch of things through, small factoring changes, stuff that just takes time to type o…

promote yourself to PM only and use agents for authoring, verification, tests, checking the tests orchestrator -> parallel subagents with investigation, authoring, verification, be…

I built a small standalone interactive to help make diffusion sampler steps more intuitive. It’s focused on the inference-time update: how the sampler moves from the current noised…

The classic papers describing diffusion are full of dense mathematical terms and equations. For many (including myself) who haven’t stretched those particular math muscles since di…

Something like a perplexity/log-likelihood measurement across a large enough number of prompts/tokens might get you the same in a statistical sense though. I expect those compariso…

AFAIK seed determinism can't really be relied upon between two machines, maybe not even between two different gpus.

Is modelwrap running on arbitrary clients? I'm not following the whole post, but how are you able to maintain confidence in client-owned hardware/disks following the secure model t…

Why does there seem to be such a divide in opinions on AI in coding? Meanwhile those who "get it" have been improving their productivity for literally years now.

> got a load of ticking time bomb bugs Lots and lots of tests!

Either really comprehensive tests (that you read) or read it. Usually i find you can skim most of it, but like in core sections like billing or something you gotta really review it…

AI is getting to the game-changing point. We need more hand-written reflections on how individuals are managing to get productivity gains for real (not a vibe coded app) software e…

I'm not too sure about this take. The larger code rewrite issue is constantly trying to be solved, which is somehow making the problem worse. In another view, standard libraries do…

I'm pretty sure it's called "reading the code". That said, it is difficult enough in its own right.

This generalization issue in RL in specific was detailed by OpenAI in 2018 https://arxiv.org/pdf/1804.03720

Recently, I've been using a local docker container to house the interpreter for all of my new python projects. For day-to-day work it is far superior than endless virtualenvs clogg…

Note that I had to remove two of the test cases to fit in the HN character limit: { name: "Large Input Slice", input: []any{"A", "B", "C", "D", "E", "F"}, chunkSize: 3, expectedChk…

Recently, I've been working on a new project in golang, which I haven't worked with in a couple years. I have almost completely replaced the time I usually spend searching stackove…