MATTEHWHOU

3 ms·

1. 1.▲
   by MATTEHWHOU 8mo ago

   My current workflow: I describe the component in plain English with specific constraints ("a data table with sortable columns, sticky header, and virtual scrolling for 10k+ rows"), let the LLM generate the first pass, then manuall
2. 2.▲
   by MATTEHWHOU 8mo ago

   There's a version of this argument I agree with and one I don't. Agree: if you use AI as a replacement for thinking, your output converges to the mean. Everything sounds the same because it's all drawn from the same distribut
3. 3.▲
   by MATTEHWHOU 8mo ago

   This is one of those projects that sounds impossible until you realize CUDA is basically C++ with some extensions and a runtime library. The hard part isn't the language translation — it's matching NVIDIA's highly optimized l
4. 4.▲
   by MATTEHWHOU 8mo ago

   I've been A/B testing the big three (GPT-5, Claude Opus 4, Gemini 3.1) on a real codebase migration this week. Quick take: Gemini 3.1 Pro's long context is genuinely better now — I fed it a 200k token codebase and it could re
5. 5.▲
   by MATTEHWHOU 8mo ago

   The interesting thing about llms.txt isn't the file format — it's the incentive shift. With robots.txt, you were telling crawlers to go away. With llms.txt, you're inviting them in and curating what they see. That's a fu