HackerTrans
TopNewTrendsCommentsPastAskShowJobs

martin_

no profile record

Submissions

Anthropic Launches Claude Desktop Extensions

anthropic.com
5 points·by martin_·vorig jaar·0 comments

Monitoring Claude Code with OTel / Datadog

ma.rtin.so
1 points·by martin_·vorig jaar·2 comments

[untitled]

1 points·by martin_·vorig jaar·0 comments

Make Something Heavy

workingtheorys.com
1 points·by martin_·vorig jaar·0 comments

Show HN: Summarize, RAG Search, and Share Podcasts - Audio Brevity

audiobrevity.com
4 points·by martin_·vorig jaar·0 comments

[untitled]

1 points·by martin_·2 jaar geleden·0 comments

comments

martin_
·4 maanden geleden·discuss
Oops sorry just saw this - done! https://unitedstarlinktracker.com/#airports
martin_
·4 maanden geleden·discuss
I built something similar[0] a while back, Stardrift looks 100x better - nice work!

[0] unitedstarlinktracker.com
martin_
·vorig jaar·discuss
how do you low cost run a 1T param model?
martin_
·vorig jaar·discuss
You can get the un-redacted prompts via the OTel event logger[0], but unfortunately it won't give you responses. You could open up a GitHub issue to request that addition!

Disclaimer - I work at Anthropic but not on Claude Code, the team is responsive via GH issues though!

[0] https://docs.anthropic.com/en/docs/claude-code/monitoring-us...
martin_
·vorig jaar·discuss
nice ship! I wrote a blog post on how to observe trends over time for a team via OTel, but I prefer your method for individual development!

https://ma.rtin.so/posts/monitoring-claude-code-with-datadog...
martin_
·vorig jaar·discuss
huh - really? no - definitely not. haven't heard anyone else report that either. What browser?
martin_
·vorig jaar·discuss
Amazing performance! Do you anticipate making the model available for commercial use or are you primarily focused on releasing agents built upon it?
martin_
·vorig jaar·discuss
Wow brutal roasts

“You've spent so much time reverse engineering other people's APIs that you forgot to build something people would want to reverse engineer.”
martin_
·vorig jaar·discuss
i have a rudimentary pipeline that takes in a ton of data, converts to json & markdown and then i used claude and o1 pro to generate the dashboard. That is to say, there are manual hops. How would you want it packaged / what would be useful?
martin_
·vorig jaar·discuss
This is neat - I built something in a similar vein recently but less productized, good work!

I have a habit of moving around (thus changing doctors) and a lack of a history / being able to easily consolidate them has been a reoccurring pain. I recently had a full physical while in Thailand and there was some potential concern around my labs & some imaging --- of course, without history it was a "check again in 6 months!" which prompted me to capitalize on Gemini's PDF parsing abilities...

I still have work to do, but it's amazing what you can do in just a few hours now: https://health.martinamps.com
martin_
·2 jaar geleden·discuss
I've observed given that LLM's inherently want to autocomplete, they're more inclined to keep complicating a solution than rewrite it because it was directionally bad. The most effective way i've found to combat this is to restart a session and prompt it such that it produces an efficient/optimal solution to the concrete problem... then give it the problematic code and ask it to refactor it accordingly
martin_
·2 jaar geleden·discuss
I for one am grateful for that fact ;)
martin_
·2 jaar geleden·discuss
to play devils advocate, if that person had decided to go with $0 instead that there would be equally bad headlines/interpretations of "Instead of allocating the formulaic $1 we are entitled to inline with all other changes over X years, they squandered it on Y"?