Nathan C. Frey

657 posts

Nathan C. Frey banner
Nathan C. Frey

Nathan C. Frey

@nc_frey

CTO & Co-Founder @CoefficientBio | Ex-@PrescientDesign • @Genentech | Advisor @atomscale & @guidelabsai | @MIT, @Penn PhD, @BerkeleyLab

New York, NY Katılım Nisan 2014
1.5K Takip Edilen4.1K Takipçiler
Sabitlenmiş Tweet
Nathan C. Frey
Nathan C. Frey@nc_frey·
my "how to get into BioML" post has been out long enough that multiple people at neurips told me that they read it, followed the advice, and now work in the industry! that's incredibly cool. naturally, my next question is, why aren't there more of you?? my answer is in the post below - a call to arms and playbook for biotechnologists to rally and radically transform our industry for the better.
Nathan C. Frey tweet media
English
13
16
135
9.7K
Nathan C. Frey retweetledi
Eryney
Eryney@eryney_ok·
Long shot, but I'm looking to connect with anyone -- residents or faculty -- in radiation oncology at Stanford, UCSF, Mt. Sinai and NYU Grossman. Would appreciate a boost, thanks!
English
4
15
40
10.9K
Nathan C. Frey
Nathan C. Frey@nc_frey·
Join the early team @CoefficientBio. We have an AI team that I’m incredibly proud of, and truly think is the most effective and exceptional technical team building in AI x bio. We're looking for people to: * Build and maintain robust AI systems to an exacting standard * Design and run experiments that let us iterate fast * Work on challenging scientific and engineering problems for human flourishing NYC-based (or willing to relocate). If you’re curious about what we’re building, reach out directly. If we don’t know each other yet, a great cold DM (or email: join@coefficientbio.com), or warm intro goes a long way.
English
12
34
328
41.4K
Nathan C. Frey
Nathan C. Frey@nc_frey·
@shelbynewsad and @RuxandraTeslo are bio guests I'd want to hear @dwarkesh_sp interview. who else?
Dwarkesh Patel@dwarkesh_sp

My main bottleneck is finding excellent guests. So, I’m hiring a couple scouts to help me find the next David Reich/Sarah Paine/Adam Brown. $100/hour, remote, I expect it’ll be 5-10 hours a week. Ideal candidate is maybe a grad student/post doc/or working in one of the fields I wanna find guests in. I'm looking to hire ~one scout for each of the following fields: bio, history, econ, math/physics, AI. The application form is linked in the essay below, and it’s extremely simple - just pitch me on a guest and tell me a bit about yourself. Deadline Fri, Jan 23. The essay includes my thoughts about what makes for an excellent guest. And I also listed a bunch of questions I'm curious about right now (to help brainstorm guests).

English
1
1
18
3.3K
Nathan C. Frey
Nathan C. Frey@nc_frey·
the new generation of AI x Bio talent are lucky to have @owl_posting primers on JPM. it used to take years of experience and deep industry connections to learn this stuff. most TechBio entrepreneurs get laughed out of dozens of hotel room meetings before learning the truth. "the entirety of California is built on top of one immensely large organism, and the particular spot in which the Westin St. Francis Hotel stands—335 Powell Street, San Francisco, 94102—is located directly above its beating heart". open.substack.com/pub/abhishaike…
Nathan C. Frey tweet media
English
2
10
95
7.7K
Nathan C. Frey retweetledi
Nitya Sridhar
Nitya Sridhar@nityasnotes·
It is extremely underrated how fun and easy it is to tinker with software vs years of knowledge building required to work with real ideas in biotech
English
2
4
40
3.8K
Nathan C. Frey
Nathan C. Frey@nc_frey·
this is the most exciting AI for science result of the year, once you think about what this means for research. "we developed a method for using ego-centric data from humans to improve our models, providing a roughly 2x improvement on tasks where robot data is limited" rooting for @physical_int and @svlevine !
Sergey Levine@svlevine

It turns out that VLAs learn to align human and robot behavior as we scale up pre-training with more robot data. In our new study at Physical Intelligence, we explored this "emergent" human-robot alignment and found that we could add human videos without any transfer learning!

English
4
8
135
15.7K
Nathan C. Frey retweetledi
Eric Nguyen
Eric Nguyen@exnx·
We’re looking for a comp bio intern to work on next-gen, multimodal bio foundation models at Radical Numerics Our team designed & trained Evo and Evo 2. Now we’re going far beyond DNA Looking for experience w/ multi-omic data, NGS, and pipelines, and on-site in Palo Alto If you want to work on science moonshots and have experience w/ multi-omic data, pipelines, reach out
English
11
30
242
28.1K
Nathan C. Frey
Nathan C. Frey@nc_frey·
my "how to get into BioML" post has been out long enough that multiple people at neurips told me that they read it, followed the advice, and now work in the industry! that's incredibly cool. naturally, my next question is, why aren't there more of you?? my answer is in the post below - a call to arms and playbook for biotechnologists to rally and radically transform our industry for the better.
Nathan C. Frey tweet media
English
13
16
135
9.7K
Nathan C. Frey retweetledi
Kyunghyun Cho
Kyunghyun Cho@kchonyc·
i gave a keynote talk at NeurIPS'25 just last week. here's the slide deck (link below) i've used to share my thoughts on who we are and what we do.
Kyunghyun Cho tweet mediaKyunghyun Cho tweet mediaKyunghyun Cho tweet mediaKyunghyun Cho tweet media
English
3
28
249
20.2K
Nathan C. Frey
Nathan C. Frey@nc_frey·
thanks to the GOAT @kchonyc for the shoutout during his @NeurIPSConf keynote address! physicists and computer scientists have a natural alliance because our disciplines are fundamentally about breaking down problems and coming up with systematic, generalizable solutions. i've never seen a more cogent explanation of this than in @kchonyc 's keynote talk. i'm biased, but you really should watch it if you want to learn from the best about problem selection and problem solving.
Nathan C. Frey tweet mediaNathan C. Frey tweet media
English
0
4
49
4.1K
Nathan C. Frey
Nathan C. Frey@nc_frey·
I am @NeurIPSConf and going to 4 events tonight and 2 tomorrow (wish me luck). if we didn't schedule time or I didn't get to your message, find me and say hello! I live in NYC, but I'm from the Midwest so I'm very friendly and approachable. And you can always lead with a "May I meet you?"
English
1
5
88
20.6K
Nathan C. Frey retweetledi
Chaitanya K. Joshi
Chaitanya K. Joshi@chaitjo·
Introducing gRNAde: our own little "AlphaGo Moment" for RNA design! 🧬🚀 📝: tinyurl.com/gRNAde-paper Unlike proteins, RNA design has long relied on "wisdom of the crowd" (human experts) or the slow crawl of directed evolution — gRNAde changes that! 🧵👇
Chaitanya K. Joshi tweet media
English
10
47
225
32.1K
Nathan C. Frey
Nathan C. Frey@nc_frey·
@juliusadml, @asalam_91, and the team @guidelabsai are building Interpretable Intelligence and launching the first 8B parameter language model grounded in human-understandable concepts, where any output can be traced back to the training data that produced it. If a model makes a clinical diagnosis or rejects a job applicant, you can see why - which training examples shaped the output, and which concepts the model is using. Guide pioneered 4 major research advances that underpin Interpretable Intelligence: 1. Atlas, an automated system for annotating training data with human-interpretable concepts 2. A new discrete diffusion architecture using block causal attention scaled to billions of parameters. 3. An 8B parameter model with no interpretability-performance tradeoff. 4. PRISM, a suite of smaller models that efficiently perform training data attributions. This is the first time that large-scale language models have achieved inherent interpretability without degrading performance. Guide has demonstrated that understanding, reliability, and auditability can be fundamental properties of models.
Guide Labs@guidelabsai

Excited to announce our $9M USD seed round led by @Initialized and the first large-scale interpretable LLM: an 8 billion parameter model capable of explaining its outputs through mechanisms humans can actually understand.

English
4
1
23
1.5K
Nathan C. Frey retweetledi
Julius Adebayo
Julius Adebayo@juliusadml·
We are doing a tutorial on interpretability @NeurIPSConf this year. What if interpretability wasn't post-hoc, but built into an LLM's architecture and pre-training? Can it scale to billions of parameters? Is there really an interpretability–performance tradeoff? Can we get faithful and reliable training-data attribution in a single forward pass? Come to the tutorial on Tuesday at 9am to hear what we’ve found. The @GuideLabsAI team will be in San Diego, all week, with more results to share.
Shichang (Ray) Zhang@ShichangZhang

Excited to share our @NeurIPSConf 2025 tutorial on Explainable AI. 📅 Tuesday Dec 2, 9:30 AM – 12:00 PM 📍 Exhibit Hall G,H, San Diego Convention Center Together with @hima_lakkaraju and @juliusadml, we cover the full landscape of explainable AI. 🔗 shichangzh.github.io/xaiTutorial/

English
2
10
81
11K
Nathan C. Frey retweetledi
Amy Lu
Amy Lu@amyxlu·
For women / under-represented minorities applying to AI PhD programs this year: if you’d like feedback on your application, feel free to consider me a resource, particularly if you’re a non-traditional applicant. DMs open! (Paying it forward for the @WiMLworkshop community 💛)
English
7
35
294
23.8K

Keşfet

@CoefficientBio @Ginkgo @shelbynewsad @RuxandraTeslo @dwarkesh_sp @owl_posting @physical_int @svlevine