Viewing profile — bhaktatejas922

thanks!

The most common failures for production agents are behavioral: looping, reasoning leakage, user frustration, and more. Using a frontier model like GPT or Sonnet to judge every turn…

do you think the industry is overfixated on startup times? what are better metrics people building with sandboxes should pay attention to

have loved simon wilson for a long long time and still do. These patterns are all out of date by at least a year - the best devs I know were using Claude 3.5 like this

100%

unfortunately, this is the reality we're in today. I see it at every scale. this is a swing at helping.

its more of a comment on human nature. on average when we did monitoring at scale reviewers averaged around 150 lines of code actually reviewed. thats the normal, day to day human …

woops this was a mistake I just made. fixed it - shouldnt require any auth

I review PRs all day and I've basically stopped reading them. Someone opens a 2000-line PR, I scroll, see it's mostly AI-generated React components, leave a comment, merge. I felt …

Hello HN, We’re the team behind WarpGrep. It’s a FAST context retrieval subagent designed to fix coding agents spending ~60% of their time searching for context + the huge context …

this. esp 32 supermini is $3 and has wifi and bluetooth. Arduino stopped being useful many years ago

https://morphllm.com/benchmarks Using fast apply is more reliable at first pass and is faster/cheaper. You prompt your agent to output in a lazy format and our model learns hoe to …

We just doubled our speculative code edit throughput and hit 10k tok/sec per request! Morph now merges code at 10,500 tokens/sec — roughly 4× faster than the best speeds on Cerebra…