Hacker News: rfurmani

Windsurf employee #2: I was given a payout of only 1% what my shares where worth

rfurmani — Thu, 24 Jul 2025 17:15:41 +0000

Article URL: https://twitter.com/premqnair/status/1948420769945682413

Comments URL: https://news.ycombinator.com/item?id=44673296

Points: 672

# Comments: 525

New comment by rfurmani in "Gemini with Deep Think officially achieves gold-medal standard at the IMO"

rfurmani — Mon, 21 Jul 2025 17:59:38 +0000

Dollars of compute at market rate is what I'd like to see, to check whether calling this tool would cost $100 or $100,000

New comment by rfurmani in "Gemini with Deep Think achieves gold-medal standard at the IMO"

rfurmani — Mon, 21 Jul 2025 17:56:57 +0000

Sounds like it did not:

> This year, our advanced Gemini model operated end-to-end in natural language, producing rigorous mathematical proofs directly from the official problem descriptions – all within the 4.5-hour competition time limit

GoogleDeepMind announces official IMO Gold

rfurmani — Mon, 21 Jul 2025 16:51:10 +0000

Article URL: https://twitter.com/lmthang/status/1947334216544621041

Comments URL: https://news.ycombinator.com/item?id=44637387

Points: 1

# Comments: 0

Show HN: I couldn't find a good AI tool for my research, so I built one

rfurmani — Wed, 25 Jun 2025 16:02:52 +0000

I went back to some pure math research recently and was looking for tools to help with the exploratory phase: navigating the literature, connecting ideas between fields with very different terminology, and tracking thoughts and approaches long term. I felt there were a lot of gaps, so I spent the last few months training models and building my own solution, Sugaku.net.

It started with a focus on mathematics, and after a lot of great feedback from professors, I've expanded it, ingesting and indexing over 200 million papers from across all academic fields. The support for non-math fields is newer and more experimental, so I'd love your feedback on it.

What it does:

- Ask questions: You can ask natural language questions against the entire 200M paper corpus.

- Explore connections: citation prediction and semantic similarity to discover work you might have missed.

- Build Projects: Organize papers, notes, and ideas into projects that you can share with others. You can use an AI collaborator within the project to brainstorm or summarize.

- Stay Updated: It actively monitors new preprints and papers to alert you to research relevant to your interests.

The site is https://sugaku.net/

I'd love your feedback, especially from those of you outside of mathematics, on the search and the workflow. Also happy to get any feature requests.

Comments URL: https://news.ycombinator.com/item?id=44378795

Points: 2

# Comments: 0

New comment by rfurmani in "Solving LinkedIn Queens Using Haskell"

rfurmani — Tue, 24 Jun 2025 15:29:26 +0000

Only for that particular board, in general it will be very complex and depend on the shape of the colored regions

New comment by rfurmani in "Waymo rides cost more than Uber or Lyft and people are paying anyway"

rfurmani — Sun, 15 Jun 2025 03:40:49 +0000

I've had a couple bad experiences with Lyft recently, including one time the driver must have clicked that they picked me up while a block away, because I could see the lyft driving to the destination without me. I tried to get a refund since I was obviously waiting my start location the whole time, but the system claimed the drive went from start to finish (even though I wasn't in the car), so no refund.

Riemann Hypothesis Project Phase 1

rfurmani — Wed, 14 May 2025 16:36:00 +0000

Article URL: https://sugaku.net/projects/76/

Comments URL: https://news.ycombinator.com/item?id=43986519

Points: 2

# Comments: 0

New comment by rfurmani in "The End of Sierra as We Knew It, Part 1: The Acquisition"

rfurmani — Fri, 04 Apr 2025 19:51:18 +0000

Completely agree on both counts! I loved those two games and felt Conquests of the Longbow didn't get the recognition it deserves.

On the second point, when I read his book (https://kensbook.com/) I was disappointed to not hear about the magic of the games themselves and the creative process behind them. It became clear that his primary goal was to grow a business, he thought being a game distributor was more exciting, but then was disrupted by Steam, shareware, and online distribution.

New comment by rfurmani in "PaperBench"

rfurmani — Wed, 02 Apr 2025 21:17:11 +0000

I'm building such tools at https://sugaku.net, right now there's chatting with a paper and browsing similar papers. Generally arXiv and other repositories want you to link to them and not embed their papers, which makes it hard to build inline reading tools, but it's on my roadmap to support that for uploaded papers. Would love to hear if you have some feature requests there

Can LLMs help with math research questions?

rfurmani — Sat, 29 Mar 2025 16:46:25 +0000

Article URL: https://sugaku.net/content/ai-benchmarks-vs-real-math-research/

Comments URL: https://news.ycombinator.com/item?id=43516794

Points: 2

# Comments: 0

New comment by rfurmani in "Devs say AI crawlers dominate traffic, forcing blocks on entire countries"

rfurmani — Wed, 26 Mar 2025 04:40:12 +0000

I've wanted to but wasn't sure how to keep track of individuals. What works for you? IP Addresses, cookies, something else?

New comment by rfurmani in "Devs say AI crawlers dominate traffic, forcing blocks on entire countries"

rfurmani — Wed, 26 Mar 2025 03:57:39 +0000

After I opened up https://sugaku.net to be usable without login, it was astounding how quickly the crawlers started. I'd like the site to be accessible to all, but I've had to restrict most of the dynamic features to logged in users, restrict robots.txt, use cloudflare to block AI crawlers and bad bots, and I'm still getting ~1M automated requests per day (compared to ~1K organic), so I think I'll need to restrict the site to logged in users soon.

The Disconnect Between AI Benchmarks and Math Research

rfurmani — Tue, 25 Mar 2025 16:05:02 +0000

Article URL: https://sugaku.net/content/ai-benchmarks-vs-real-math-research/

Comments URL: https://news.ycombinator.com/item?id=43472897

Points: 4

# Comments: 0

The cultural divide between mathematics and AI

rfurmani — Wed, 12 Mar 2025 16:07:35 +0000

Article URL: https://sugaku.net/content/understanding-the-cultural-divide-between-mathematics-and-ai/

Comments URL: https://news.ycombinator.com/item?id=43344703

Points: 300

# Comments: 177

New comment by rfurmani in "Hard problems that reduce to document ranking"

rfurmani — Tue, 25 Feb 2025 18:36:52 +0000

Very cool! This is also one of my beliefs in building tools for research, that if you can solve the problem of predicting and ranking the top references for a given idea, then you've learned to understand a lot about problem solving and decomposing problems into their ingredients. I've been pleasantly surprised by how well LLMs can rank relevance, compared to supervised training of a relevancy score. I'll read the linked paper (shameless plug, here it is on my research tools site: https://sugaku.net/oa/W4401043313/)

New comment by rfurmani in "Show HN: Tools for Math Research"

rfurmani — Mon, 17 Feb 2025 19:02:11 +0000

That's fair, though there's not much to say since I'm building it out myself as a benefit corporation. I also have strict opt-out for any communications and a proper privacy policy.

I've also tried to keep as much as I can accessible without login, but I want to protect some of the more expensive features from being spammed.

Without signup up you can:

* explore works (but not chat with them) https://sugaku.net/oa/W4206400500/

* explore authors https://sugaku.net/oa/A5059543195/

* see and share AI answers (eg https://sugaku.net/qna/4e59662a-a938-404e-8c0b-b9dc79e37c29/ and https://sugaku.net/qna/517930ff-42ad-47c5-9d9c-e807d06a8453/)

* prompt for new paper ideas https://sugaku.net/current/papergen/

* see and share these ideas https://sugaku.net/current/papergen/idea/719aed36-8dcd-4fd1-...

New comment by rfurmani in "Show HN: Tools for Math Research"

rfurmani — Mon, 17 Feb 2025 18:55:02 +0000

The flipside of that is seeing hints of a result that would be really helpful. I still remember how excited I was to stumble on a book from 1931 (The Taylor Series by Dienes) since it had the only english-language proofs of some results by Szego and Polya that I felt could unblock my research. My hope is that this discovery problem can be largely solved.

This is also why I'm not as excited by the focus on pure reasoning and olympiad problem solving in the math and AI space. It's like the early career phase of trying to solve Collatz and Riemann but just repeating work from decades ago.

New comment by rfurmani in "Caddy – The Ultimate Server with Automatic HTTPS"

rfurmani — Sun, 16 Feb 2025 19:46:03 +0000

I'm serving AI models on Lambda Labs and after some trial and error I found having a single vllm server along with caddy, behind cloudflare dns, to work really well and really easy to set up

vllm serve ${MODEL_REPO} --dtype auto --api-key $HF_TOKEN --guided-decoding-backend outlines --disable-fastapi-docs &

sudo caddy reverse-proxy --from ${SUBDOMAIN}.sugaku.net --to localhost:8000 &

New comment by rfurmani in "Show HN: Tools for Math Research"

rfurmani — Sat, 15 Feb 2025 00:15:09 +0000

As a former mathematician, I found research to be a very winding path. While that can be fun, I felt there's a lot of opportunity to train LLMs and ML models on the corpus of math papers, to try to make research more deliberate and less reliant on talking to the right person at the right time.

This is very much a work in progress but so far you can:

* Browse through similar papers

* Get recommendations for new papers and collaborators

* Chat with papers and ask questions to all the major reasoning models

* Have it come up with future paper ideas (along with references) giving a potential title or collaborators.

My focus very much is on the exploratory stages since that's where a lot of the time is spent, but I intend to integrate more tools for problem solving, writing, and computation.