iiinova — where your digital self keeps evolving

Reposted by Jesse G Meyer PhD

Ben Neely @benneely.com · 21/07/2026

Self-promotion or PSA, a little of both? Our (@magnuspalmblad.bsky.social and Jared Ragland) Letter is out in JPR, which tackled the age old question "how can we make AIs follow community-supported domain knowledge?" Answer: use a GROUNDING.md file in your context. 1/7 pubs.acs.org/doi/10.1021/...

pubs.acs.org

Agentic AI-Assisted Coding Offers a Unique Opportunity to Instill Epistemic Grounding during Software Development

The capabilities of AI-assisted coding are progressing at a breakneck speed. Chat-based vibe coding has evolved into fully fledged AI-assisted, agentic software development using agent scaffolds, wher...

193

Reposted by Jesse G Meyer PhD

Science X / Phys.org @sciencex.bsky.social · 27/06/2026

A hospital AI model identified patients at risk of low blood sugar up to 24 hours before it occurred. It used routine health record data tracked in four-hour intervals over five days. doi.org/hb8xbs

medicalxpress.com

Hospital AI tool predicts low blood sugar in patients up to 24 hours in advance

Cedars-Sinai Health Sciences University investigators developed an AI-based model that can identify hospitalized patients at risk of low blood sugar up to 24 hours before the condition occurs.

021

Jesse G Meyer PhD @j-my-sci.bsky.social · 27/06/2026

Just text. Figures would be better but I had to keep everything super simple in order to make it cheap enough that I can afford to run >5,000 preprints through AI every month. Currently costs me about $45 in gemini tokens plus $20 in web hosting per month. More details here unpeer.org/how-it-works

unpeer.org

Unpeer — Discover Preprints

Unpeer scores and ranks BioRxiv preprints by impact, novelty, and rigor. Search, filter, and discover the strongest preprint science.

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 26/06/2026

To everyone upset about preprint ranking (spurred by the QED 1%): let's remember the problem isn't the ranking.

021

Jesse G Meyer PhD @j-my-sci.bsky.social · 25/06/2026

Pretty wild to see the polarization about using AI to review preprints on X when I started doing that four months ago with Unpeer dot org and not many noticed. Also checking x made me remember why I deleted my x account

021

Jesse G Meyer PhD @j-my-sci.bsky.social · 15/05/2026

here is a lecture I gave to students at CalTech, completely unscripted, on some of our recent work related to proteomics and AI www.youtube.com/watch?v=Ellg...

youtube.com

Proteomics and AI

YouTube video by omics and AI

150

Jesse G Meyer PhD @j-my-sci.bsky.social · 14/05/2026

Is B- a good grade? Does it read the whole paper or just the abstract

100

Reposted by Jesse G Meyer PhD

Ben Neely @benneely.com · 14/05/2026

It got picked up by my weekly proteomics horizon scanner (1 of 5, I limit at 5). B-)

111

Jesse G Meyer PhD @j-my-sci.bsky.social · 14/05/2026

Love it! Did you say this is a gemini gem?

110

Jesse G Meyer PhD @j-my-sci.bsky.social · 08/05/2026

Chirashi? This used to be my favorite food and then I watched seaspiracy maybe 7 years ago and haven't had real sushi since

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 08/05/2026

Out of 78 publications, this new one is the most important we have ever written. Robotic perturbation proteomics and AI agents enable scalable drug mechanism discovery www.biorxiv.org/content/10.6...

biorxiv.org

160

Jesse G Meyer PhD @j-my-sci.bsky.social · 27/04/2026

It's pay walled. What shoes did he wear

200

Jesse G Meyer PhD @j-my-sci.bsky.social · 09/04/2026

A practical scale for manuscript readiness

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 08/04/2026

New feature - subscribe for free monthly emails with the preprint summaries in the 11 focus areas on unpeer.org

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

Curious to hear if there is overlap in the proteomics preprints you found versus those on Unpeer

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

Also journals are very very slow. Unpeer is fast

110

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

My ultimate goal is to build something that replaces all the bad parts about journals while keeping the reasons we have journals to start with. Slow and stochastic review replaced by consistent AI review. Keep quantitative proxy for importance for use by hiring committees or potential readers

121

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

Great question. Nearly 30,000 preprints/month. Human review is slow, inconsistent, and doesn’t scale. Journals still take our time and money. We need AI to filter. If everyone switches to unpeer for preprint curation, we don't need journals anymore. Fast, free, signal from noise

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

I spent $150 on AI to review all the >15,000 preprints posted on biorxiv @biorxivpreprint.bsky.social in 2026 so you don't have to. Results sorted into focus areas (Lenses) online free at unpeer.org Not perfect. But it scales.

110

Jesse G Meyer PhD @j-my-sci.bsky.social · 06/04/2026

Made some UI changes to unpeer.org and now you can subscribe to your monthly virtual issues of interest via email for free. Out of over 5,000 preprints that appear on @biorxivpreprint.bsky.social every month, get a quick sense of strengths and weaknesses from AI reviews, free

unpeer.org

Unpeer — Discover Preprints

Unpeer scores and ranks BioRxiv preprints by impact, novelty, and rigor. Search, filter, and discover the strongest preprint science.

000

Reposted by Jesse G Meyer PhD

PastelBio @pastelbio.bsky.social · 30/03/2026

Single‐Injection Multi‐Omics Analysis by Direct Infusion Mass Spectrometry onlinelibrary.wiley.... --- #proteomics #prot-paper

051

Reposted by Jesse G Meyer PhD

Jason Moore @moorejh.bsky.social · 26/03/2026

Vibe Coding Comes to Omics - How @j-my-sci.bsky.social from @cedarssinaihsu.bsky.social built a #proteomics data analysis application in under 10 minutes, using four prompts, no handwritten code, & under $2 www.theanalyticalscientist.com/issues/2026/... #coding #vibecoding #bioinformatics

theanalyticalscientist.com

Vibe Coding Comes to Omics

How Jesse Meyer built a fully functional proteomics data analysis application in under 10 minutes, using four prompts, no handwritten code, and under $2

351

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

It was peer reviewed. Figure 2 directly compares the vibe coded platform to a local reference pipeline using simulated data with known ground truth. That validation step was improved during review.

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

Thanks for sharing this. I’ve found that hands on use clarifies both the strengths and the limits. In experienced hands these tools can accelerate iteration, but they still require judgment and validation like anything else in science. I try to emphasize that balance in the paper too.

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

The debate around vibe coding in science is exactly what should be happening. New automation always raises real concerns. The path forward is not blind adoption or blanket rejection, but shared standards

000

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

Grateful you shared this. The intent was to spark exactly this discussion: how we can use AI to speed up tool building while keeping validation, reproducibility, and rigor front and center. I very much welcome the conversation.

030

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

Models absolutely inherit bias. That is a real concern. But this paper is not training AI to infer biology from literature. It uses an LLM as a coding assistant to scaffold a standard proteomics workflow, then validates the outputs against a reference pipeline with known ground truth.

100

Reposted by Jesse G Meyer PhD

Hiren Joshi @glyco.me · 30/01/2026

I am a fan of judicious vibe-coding, but it requires training in a methodology to evaluate results. E.g., I pointed a wet lab PhD student to a LLM to code an excel formula for decoding mass spec composition strings. I think this is OK where you have an orthogonal method to validate your results.

2101

Reposted by Jesse G Meyer PhD

Rod Page @rdmpage.bsky.social · 30/01/2026

What specifically don’t you like about the paper? It explains the approach to generating code, tests that the code produces correct results for a dataset with known properties, and provides the prompts, code, and data. The discussion includes caveats and areas needing further work.

011

Reposted by Jesse G Meyer PhD

Jase Gehring @skyjase.bsky.social · 30/01/2026

in one month i vibe coded, trained, and eval'd a new family of deep learning models for de novo peptide sequencing. I applied ideas from ML preprints from Oct and Dec 2025 and achieved comparable performance to Casanovo on a single consumer-grade GPU

231

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

I have had several related experiences. Building in days huge deep learning ideas, testing many variants, and finding quickly that it's not a worthwhile. Imagine that a PhD student had instead done that manually over years and then had nothing for their dissertation

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 16/02/2026

If you have tried vibe coding, you know it can be a super power for experienced coders 👇

010

Reposted by Jesse G Meyer PhD

Ben Neely @benneely.com · 30/01/2026

On the "vibe coding omics analysis is here" demo paper, and some responses (run for the hills!), a thread for myself: - we know that LLM-assisted or even driven coding is here. if you haven't tried it even in the last 3 months, you are behind - yes it is powerful and enabling 1/7

382

Jesse G Meyer PhD @j-my-sci.bsky.social · 31/01/2026

Thanks for reading it and for your perspective

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 30/01/2026

Most genius ideas seem obvious in hindsight

030

Jesse G Meyer PhD @j-my-sci.bsky.social · 23/01/2026

We are all software engineers now

010

Jesse G Meyer PhD @j-my-sci.bsky.social · 22/01/2026

No apologies needed! Just wait until you try the coding interfaces like Claude Code or antigravity! Antigravity will write thousands of lines for you if you give it clear long term goals and test definitions

100

Jesse G Meyer PhD @j-my-sci.bsky.social · 18/01/2026

We got the NOA for an MPI R01 from the NIA yesterday. Greatful for all the collaborators, facilitators, mentors, and trainees in my group who made this possible.

080

Reposted by Jesse G Meyer PhD

Jesse G Meyer PhD @j-my-sci.bsky.social · 09/01/2026

It's hard to stay up to date with new literature. So many papers each week, which ones should I read? That is why I built ReScoop.xyz - try it for free! If you find it useful, less than the price of one Starbucks per month

rescoop.xyz

Research Scoop - AI-Powered Research Paper Discovery

Stay ahead of research with AI-powered paper discovery. Get weekly digests of the latest papers from PubMed, arXiv, bioRxiv, and medRxiv tailored to your interests.

011

Jesse G Meyer PhD @j-my-sci.bsky.social · 11/01/2026

I haven't tried Claude Code personally but I have watched people in my group using it and it looks awesome. In my experience complex ideas are possible in antigravity if you define good test cases and go step by step

100

Jesse G Meyer PhD @j-my-sci.bsky.social · 09/01/2026

rescoop.xyz

Research Scoop - AI-Powered Research Paper Discovery

Stay ahead of research with AI-powered paper discovery. Get weekly digests of the latest papers from PubMed, arXiv, bioRxiv, and medRxiv tailored to your interests.

011

Jesse G Meyer PhD @j-my-sci.bsky.social · 09/01/2026

I suggest the free trial of antigravity, it's much better than copilot for software dev in my opinion

200