Hacker News: maxrumpf

New comment by maxrumpf in "Show HN: HelixDB – A Graph Database built on Object-storage"

maxrumpf — Wed, 10 Jun 2026 16:16:17 +0000

does it support fts/vector on edges of the graph?

New comment by maxrumpf in "Please don't spam people looking for employment. It's just cruel"

maxrumpf — Tue, 02 Jun 2026 15:38:31 +0000

are you a maintainer on npm?

Training SID-1 to beat GPT-5 at search with 1k+ QPS RL

maxrumpf — Wed, 20 May 2026 19:31:15 +0000

Article URL: https://turbopuffer.com/blog/reinforcement-learning-sid-ai

Comments URL: https://news.ycombinator.com/item?id=48212869

Points: 12

# Comments: 0

New comment by maxrumpf in "Launch HN: Voker (YC S24) – Analytics for AI Agents"

maxrumpf — Tue, 12 May 2026 17:55:47 +0000

congratulations!

New comment by maxrumpf in "Chroma Context-1: Training a Self-Editing Search Agent"

maxrumpf — Fri, 27 Mar 2026 03:16:50 +0000

We think this is a pretty sad day for research: Some context about Chroma's model. https://x.com/maxrumpf/status/2037365748973384154?s=20

New comment by maxrumpf in "Scaling compute for retrieval by 5 OOMs: SID-1 tech report"

maxrumpf — Mon, 08 Dec 2025 23:43:00 +0000

This is the tech report for a model I helped work on. I'm biased, but it turned out very well.

We essentially let the model learn to retrieve like a human would: Make a first search, read the results, and then make another. This lets the model be vastly better than pre-programmed pipelines. We test this extensively and compare against implementing this with API models (like Sonnet 4.5 and GPT-5.1). SID-1 compares favorably.

Happy to answer any questions or get feedback. First and foremost: Enjoy the read. It's much more detailed than most tech reports.

Scaling compute for retrieval by 5 OOMs: SID-1 tech report

maxrumpf — Mon, 08 Dec 2025 23:43:00 +0000

Article URL: https://www.sid.ai/research/sid-1-technical-report

Comments URL: https://news.ycombinator.com/item?id=46199274

Points: 4

# Comments: 1

JIT Coding not just compiling

maxrumpf — Thu, 22 Aug 2024 17:54:35 +0000

Article URL: https://maxrumpf.com/writing/2024-08-21-jit-coding.html

Comments URL: https://news.ycombinator.com/item?id=41322760

Points: 1

# Comments: 0

New comment by maxrumpf in "New anti-speeding system coming to Australia is dangerous"

maxrumpf — Wed, 21 Aug 2024 22:54:48 +0000

This comment section is scary. Hacker news advocating FOR nanny technology?!

New comment by maxrumpf in "N=1 software and the unreasonable effectiveness of Excel"

maxrumpf — Tue, 20 Aug 2024 22:34:53 +0000

Excel is awful in almost every way, but I just wish more software was as customizable.

I can get (even more) customization by using pandas etc., but it's usually much slower and you get much less of an intuition about the data.

N=1 software and the unreasonable effectiveness of Excel

maxrumpf — Tue, 20 Aug 2024 22:24:53 +0000

Article URL: https://maxrumpf.com/writing/2024-08-19-n-of-1-software.html

Comments URL: https://news.ycombinator.com/item?id=41304733

Points: 3

# Comments: 2

New comment by maxrumpf in "Polychromatic Pixels"

maxrumpf — Thu, 18 Jul 2024 17:21:41 +0000

I imagine color consistency will be such a pain here.

New comment by maxrumpf in "Tell HN: Ever think of applying to YC? Do it this weekend for S24"

maxrumpf — Fri, 19 Apr 2024 22:58:56 +0000

The weirdest thing people do is make up criteria that YC supposedly uses to reject people. There was such a huge diversity in our batch: From 20 y/o to 40+. Foreign, domestic. Credentialed, not credentialed. $1M rev run rate, $0 run rate. Just apply.

New comment by maxrumpf in "Mixture-of-Depths: Dynamically allocating compute in transformers"

maxrumpf — Sun, 07 Apr 2024 23:17:13 +0000

The abstract and the rest of the paper don't really match imo. It's not really allocating more to some sequences, but just introducing ~dropout. Might be different sides to the same coin, but was still a weird read.