Out of all the migrations I help teams with, Go to Rust is a bit of an outlier. Itβs not a question of βis Rust faster?β or βdoes Rust have types?β, Go already gets you most of the way there. The discussion is mostly about correctness guarβ¦
Today marks four years since I joined AWS. My last day will be Friday.\nI have to say being fired from AWS is actually a relief. There have been a lot of changes to the company since I joined in 2022, and the company I wanted to work for is no longer the same company.\nThis past year, while I was doing my best to make AWS play nice in open source communities, there were two main drivers making me unhappy with my job: organizational change and the acceleration of the focus on Generative AI.\n
321 pointsby RyeCombinator151 days ago134 comments
Large Language Model (LLM) agents demonstrate strong performance in autonomous code generation under loose specifications. However, production-grade software requires strict adherence to structural constraints, such as architectural patterns, databases, and object-relational mappings. Existing benchmarks often overlook these non-functional requirements, rewarding functionally correct but structurally arbitrary solutions. We present a systematic study evaluating how well agents handle structural constraints in multi-file backend generation. By fixing a unified API contract across 80 greenfield generation tasks and 20 feature-implementation tasks spanning eight web frameworks, we isolate the effect of structural complexity using a dual evaluation with end-to-end behavioral tests and static verifiers. Our findings reveal a phenomenon of constraint decay: as structural requirements accumulate, agent performance exhibits a substantial decline. Capable configurations lose 30 points on average in assertion pass rates from baseline to fully specified tasks, while some weaker configurations approach zero. Framework sensitivity analysis exposes significant performance disparities: agents succeed in minimal, explicit frameworks (e.g., Flask) but perform substantially worse on average in convention-heavy environments (e.g., FastAPI, Django). Finally, error analysis identifies data-layer defects (e.g., incorrect query composition and ORM runtime violations) as the leading root causes. This work highlights that jointly satisfying functional and structural requirements remains a key open challenge for coding agents.
A new study published in Nature's Humanities and Social Sciences Communications journal has confirmed what many workers have quietly hoped for: companies can
256 pointsby randycupertino150 days ago255 comments
The AI race, the future of AGI, and the inside story of OpenAI. Greg Brockman is the co-founder and President of OpenAI, the company behind ChatGPT and GPT-5. He was the first engineer at Stripe before leaving in 2015 to help start OpenAI. In this rare conversation, Greg goes inside the moments that built, and nearly β¦
211 pointsby prakashqwerty150 days ago240 comments
I have a huge appreciation for the Scheme programming language. I just seem to be unable to get it to stick in my head. This seems like a huge revelation for someone who named their blog after the Scheme textbook, but there it is. This post is the public admission I need to make, to...
A deep dive into the font of the "Dead Test" diagnostic cartridge of the C64, including an Easter egg, a look into the implementation, and, finally, some Commodore 8-bit character ROMs for download.
Aug 27th to Aug 30th Registration is all inclusive; we'll provide you with shared lodging, all your meals, snacks while we're hanging out in the evenings and some of the best socializing you've ever had. Plus we're probably going to have some great swag too :) We're looking forward to making the world a better place with you!
Grothendieck is revered in the world of math; outside of it, heβs known for his unusual life, if heβs known at all. But what were his actual mathematical contributions?
Today is a day IΒve dreaded for over a decade. At around 6:20 a.m., weΒll get my oldest son Judson up, give him a bath, put his safety harness on (donΒt worry, weΒll dress him first), and...