shipilovya
Autoselect the best AI model for any health question using HealthBench scores
Re:Autoselect the best AI model for any health question using HealthBe...
on Tuesday May 20, 2025 @05:30PM (#44046110)
Last week OpenAI released HealthBench, the most comprehensive set of evals for health to date. The top 3 scoring models all spiked on different things:
- GPT-4.1 is best when you need a straight answer - o3 is best for complex cases - Grok is best at clarifying important info (“truthseeking”)
Made this prototype mostly to understand HealthBench deeper. I will probably use it in the future products I make.
Re:Plasticlist Report – Data on plastic chemicals in Bay Area foods
on Saturday December 28, 2024 @12:22PM (#42532561)
These 41 samples came from 24 unique products
Re:Ask HN: How did you choose your cofounder? Were you right or wrong?
on Wednesday November 15, 2023 @11:06PM (#38285708)
Interesting, do you feel like your / your founder’s qualities and relationship had any impact on the success of failure of your startup? Would love to hear how things went.
Ask HN: How did you choose your cofounder? Were you right or wrong?
My friend and I are trying to figure out how people decide to cofound a company together, and how that affects whether it fails or succeeds.Tell us your story!1. If you had a bad cofounder experience- What went wrong?- What were you looking for in the other person, and what did you miss that led to a bad outcome?- Did you change how you choose cofounders now based on that experience?2. If you...
Show HN: Lethelink – MVP for an interface to soothe people with Alzheimer's
Re:First principles for working on your ideas
on Monday September 18, 2023 @05:56PM (#37562646)
Yep! I discovered it unexpectedly and found it works really well, at least so far. In hindsight it's obvious it would have worked in my case.
First principles for working on your ideas
Seminal reads on AI and computing
Ever wondered who decides how the Internet should evolve?
Some ideas on exposing the cognitive abyss between humans and language models
A tool that takes notes for your meetings without joining the call
Re:My 'copilot for mind' now has access to my journals from the last ~...
on Wednesday March 29, 2023 @10:55PM (#35367307)
I sent it to OpenAI because I felt comfortable doing it with my entries, but my guess is that the same can be done with an open model.
I want to see if it’s possible to have it extract more structured information about my beliefs, values, and thought patterns, and then reference it to non-intrusively comment on my writing.
Let me know if you’re interested in this, I saw your post on journaling and found it thoughtful.