Hacker News: rjeli

New comment by rjeli in "Adventures in Symbolic Algebra with Model Context Protocol"

rjeli — Thu, 22 May 2025 16:01:57 +0000

the implementations have a distinctly "I wrote this at a 3 AM hackathon" vibe

The LLM handles the natural language interaction and orchestration, while the computer algebra system does what it does best ... exact symbolic manipulation.

this smells like claude :D

New comment by rjeli in "UCSD: Large Language Models Pass the Turing Test"

rjeli — Wed, 02 Apr 2025 15:21:34 +0000

Author’s announcement xeet with some context and highlights: https://x.com/camrobjones/status/1907086860322480233

mirror: https://nitter.net/camrobjones/status/1907086860322480233#m

They link to the webapp which you can play yourself!

https://turingtest.live/

(I have a dozen games played and 100% success rate :3)

New comment by rjeli in "Google calls Gemma 3 the most powerful AI model you can run on one GPU"

rjeli — Thu, 20 Mar 2025 22:59:11 +0000

they come out thursday nights (in the US) - 9pm EST

New comment by rjeli in "Apple M3 Ultra"

rjeli — Wed, 05 Mar 2025 15:56:33 +0000

Wow, incredible. I told myself I’d stop waffling and just buy the next 800gb/s mini or studio to come out, so I guess I’m getting this.

Not sure how much storage to get. I was floating the idea of getting less storage, and hooking it up to a TB5 NAS array of 2.5” SSDs, 10-20tb for models + datasets + my media library would be nice. Any recommendations for the best enclosure for that?

New comment by rjeli in "When imperfect systems are good: Bluesky's lossy timelines"

rjeli — Thu, 20 Feb 2025 04:57:27 +0000

there are a lot of interesting people on twitter

New comment by rjeli in "Blue Origin New Glenn Mission NG-1 (video)"

rjeli — Thu, 16 Jan 2025 08:10:27 +0000

IIRC when I played KSP it was necessary to point slightly down if you wanted to reach orbit in a continuous burn, rather than waiting to burn more at perigee. Is that true in general?

(was playing with a mod that models ullage, so relighting was quite finicky)

New comment by rjeli in "Reciprocal Approximation with 1 Subtraction"

rjeli — Tue, 31 Dec 2024 18:58:18 +0000

For C++, `bit_cast(0.f)` should be Well Defined, right? I'm curious, in C, is union-casting float->uint32_t also Perfectly Legal And Well Defined?

(I am not a C or C++ expert.)

New comment by rjeli in "Prover-Verifier Games improve legibility of language model outputs"

rjeli — Wed, 17 Jul 2024 18:00:11 +0000

Funny that when I reached the "Key Findings" section, my brain immediately parsed it as ChatGPT output. Maybe it's the bullet points, the word choice, or just the font...

New comment by rjeli in "Convolutions, Fast Fourier Transform and polynomials (2022)"

rjeli — Mon, 01 Jul 2024 15:17:18 +0000

Note that the FFT has this property of “convolution is pointwise multiplication” for any cyclic multiplicative group, see https://www.sciencedirect.com/science/article/pii/S002200007... for a more algebraic derivation.

Some call this a “harmonic” fft, and there are also non-harmonic FFTs:

- the “additive NTT” of [LCH14] on GF(2^n)

- the circle fft on the unit circle X^2+Y^2=1 of a finite field [HLP24]

- the ecfft on a sequence of elliptic curve isogenies [BCKL21]

[LCH14]: https://arxiv.org/abs/1404.3458

[HLP24]: https://eprint.iacr.org/2024/278

[BCKL21]: https://arxiv.org/pdf/2107.08473

New comment by rjeli in "Group actions and hashing unordered multisets (2021)"

rjeli — Wed, 26 Jun 2024 15:57:03 +0000

There’s a nice writeup on group hashes here: https://cronokirby.com/posts/2021/07/on_multi_set_hashing/

In particular, if you choose a group where discrete log is hard (such as prime order elliptic curves), multiset hashing falls out for free

New comment by rjeli in "London Underground hosts tests for 'quantum compass' that could replace GPS"

rjeli — Mon, 17 Jun 2024 21:48:58 +0000

With high enough accelerometer accuracy, you need to start adjusting your subtracted gravity vector with maps, right?

New comment by rjeli in "Show HN: Revideo – Create Videos with Code"

rjeli — Tue, 11 Jun 2024 18:20:17 +0000

Very cool! I assume it uses WebCodec VideoEncoder to encode in browser, maybe with a wasm ffmpeg fallback? How reliable/easy to use have you found that?

New comment by rjeli in "Private Cloud Compute: A new frontier for AI privacy in the cloud"

rjeli — Tue, 11 Jun 2024 03:21:48 +0000

ZKML is actually not horrible, probably only 100-1000x overhead atm. Unfortunately it doesn’t solve the problem, you would need FHE which has much higher overhead

New comment by rjeli in "WWDC 2024 [video]"

rjeli — Mon, 10 Jun 2024 17:11:26 +0000

Got in late, seems like I can't rewind on this stream??

New comment by rjeli in "Viagra improves brain blood flow and could help to prevent dementia"

rjeli — Sun, 09 Jun 2024 17:54:07 +0000

See also levomethamphetamine, the mirror image of meth found commonly in drugstores in Vick’s inhalers, acting as a vasoconstrictor

New comment by rjeli in "mactop"

rjeli — Wed, 22 May 2024 17:42:05 +0000

Does anyone have recs for doing really low level performance tuning on Apple Silicon? Xcode Instruments isn’t very easy to use or interpret. Things like what % of time is spent on memory stalls, simd units, etc

New comment by rjeli in "Bitcoin's Future Depends on a Handful of Mysterious Coders"

rjeli — Tue, 07 May 2024 13:19:13 +0000

I’ve always thought it was strange that btc community encourages only running core, rather than having several implementations to make consensus more robust

New comment by rjeli in "Manta Ray UUV prototype completes in-water testing"

rjeli — Sat, 04 May 2024 21:16:28 +0000

hm yeah i wonder if you could use the phase change tanks as an acoustic antenna to communicate with buoys on the surface

New comment by rjeli in "Manta Ray UUV prototype completes in-water testing"

rjeli — Sat, 04 May 2024 03:34:02 +0000

You can do it without a TEG: https://sci-hub.se/10.1109/48.972077

Essentially you use a fluid which freezes at the bottom of the ocean and loses volume (some kind of paraffin)

And then you hook it up with some actuated valves and N2 accumulators so the volume change and energy harvesting happens "out of phase" - at the surface, you can deflate an external bladder and sink, and at the bottom, you use some stored energy to reinflate the bladder

With a large enough volume you get near infinite range, only have to pay for payload and any onboard electronics

this report has some more info on the absurd efficiency of thermal gliders: https://escholarship.org/content/qt1c28t6bb/qt1c28t6bb_noSpl...

New comment by rjeli in "Swan – A Lightweight Language Model Execution Environment Using FPGA"

rjeli — Fri, 26 Apr 2024 02:37:52 +0000

I'm curious, do you have any intuition for what percent of the time is spent shifting & masking vs. adding & subtracting (int32s I think)? Probably about the same?