Live data from Hacker News

Viewing profile — mattlangston

mattlangston

HN member
Joined
Fri, May 20, 2022, 6:22 AM UTC
HN karma
137
Public activity
20 items

About mattlangston

PhD physicist and Research Software Engineer. Applying experimental physics to understand how LLMs work.

https://x.com/mattlangston

https://mattlangston.com

Matt Langston holds a PhD in physics. As an experimental physicist at the Stanford Linear Accelerator Center (SLAC), he designed and built particle detectors to probe the foundations of physics at the quantum level.

He is also a Research Software Engineer with over 20 years of experience bridging research teams and software engineering teams - currently at Apple for over 10 years, and previously at SLAC, Microsoft Research, and several acquired startups.

Matt's current research applies the experimental physicist's toolkit to Large Language Models: measuring their equations of motion in search of an LLM Standard Model - a unified framework describing how and why LLMs work.

Recent public activity

  1. comment
    Comment #48514018

    Crypto Wars v2. I know how this movie ends.

  2. comment
    Comment #48044935

    Software engineering is software engineering. An ace software engineer is not an ace because of tooling. It's not the plane, it's the pilot, or something like that.

  3. comment
    Comment #47590726

    Boris Cherny has said that Claude Code is simply a client of the public Claude API, so this may be a good thing for Anthropic to demonstrate Claude API best practices. Maybe CC "le…

  4. comment
    Comment #47302911

    I had the same SETI inspiration yesterday of "autoresearch swarms". Andrej and I had no coordination - we just both had the same idea at the same time. Experiments already running.…

  5. comment
    Comment #47218660

    The future is bright for software engineers. The big takeaway isn't reverse engineering the ANE per se, but what Manjeet could do with his software engineering skills when accelera…

  6. comment
    Comment #47021268

    This is a duplicate of https://news.ycombinator.com/item?id=47005565 . Please have all conversations there about Dwarkesh Patel's interview with Dario Amodei.

  7. comment
    Comment #47010793

    Dwarkesh Patel interview with Dario Amodei (Anthropic CEO)

  8. story
  9. comment
    Comment #47003616

    The frontier-of-knowledge point is the right question. My own research is a case in point - I apply experimental physics methods to LLMs, measuring their equations of motion in sea…

  10. comment
    Comment #46986565

    PhD physicist (Stanford/SLAC), Research Software Engineer doing low-level systems work in C/C++ and LLM research. Not a founder or investor — just a practitioner. One data point fo…

  11. comment
    Comment #46986186

    Good question. It's mostly the third — but let me unpack that. The Standard Model predicts a specific value for the weak mixing angle, which determines the electron's vector coupli…

  12. comment
    Comment #46965679

    Fair - that sounds hyperbolic. But my point is specific: if the weak mixing angle is shifted from the Standard Model value, one of the standard explanations is a heavier cousin of …

  13. comment
    Comment #46954434

    Experimental particle physicist here. It's just hard. I measured the electron's vector coupling to the Z boson at SLAC in the late 1990s, and the answer from that measurement is: w…

  14. comment
    Comment #45421409

    It does well with screenshot-calculus for me. For example, I pasted a screenshot of the Layer Norm equation into Claude Code 2 and asked: "Differentiate y(x) w.r.t x, gamma and bet…

  15. story
  16. comment
    Comment #44742327

    fwiw I live in [macOS emacs]( https://emacsformacosx.com/ ) all day long for systems engineering (C/C++) and have 201 open buffers, an uptime of 57 days and ~540 MB memory usage.

  17. comment
    Comment #44718064

    I have both a Claude subscription plan and console credits as my backup, which I thought was a reasonable solution.

  18. comment
    Comment #44352751

    According to the published paper (linked in the article) Kletetschka's theory correctly predicts several experimentally measured quantities of the Standard Model. The two that jump…

  19. comment
    Comment #44333586

    "... Miyazaki warns against the dangers of harmful technology and the moral implications of its use." Apropos for current times.

  20. comment
    Comment #44313011

    Very nice find @pudiklubi. Thank you.