Alex Cabrera

@BEBischof if only I was consistent enough w/ linear tickets... HERE BE DRAGONS github.com/cabreraalex/da…

@a_a_cabrera Check out my linear tracker. This gh mini ui is great and now I want one btw. x.com/bebischof/stat…

This is a fun idea. I hate the GH email and UI updates for comments, review requests, etc. Took 30 min to get a nice toolbar app w/ your PRs and reviews, read status, etc.!

the big labs should release alexa/google home competitors. make plugins/mcps for home devices. it can remind you if you forgot to program the coffee, update the lights dynamically based on your calendar, etc.

1/ Most AI features die in demo-land. The teams that actually ship have a system: (1) evals (2) build vs buy (on repeat) (3) org design amplifypartners.com/blog-posts/how…

I wanted to share something I built over the last few weeks: isometric.nyc is a massive isometric pixel art map of NYC, built with nano banana and coding agents. I didn't write a single line of code.

For a decade, techbio optimized the drug discovery "paper mill": more molecules, faster. But the hard part - knowing which ones will work safely in humans - hardly moved at all. Late last year, I spent time with Axiom Bio, a company built around the idea that the real bottleneck is predicting clinical activity in molecules, *not* merely discovering them.

Announcing Worktrunk! A git worktree manager, designed for running AI agents in parallel. A few points on why I'm so excited about the project, and why I hope it becomes broadly adopted 🧵

You asked for it. A lot. So we built it.

1/ Some pundits are predicting that the AI bubble will burst. I doubt it. But more ideas or compute won't unlock an "intelligence explosion." The biggest bottleneck AI research faces is the pace and quality of experimentation.

if you want to write core infrastructure code for very cool ai + bio plz dm me

To replace animal testing with AI, we need MASSIVE human datasets. Today, we're thrilled to share Axiom's new data exploration tool, providing the ability to visually explore the world's largest primary human liver toxicity dataset. Built with Axiom's proprietary wetlab protocols, our dataset includes detailed liver toxicity profiles for over 100,000 distinct molecules. The key to this dataset is our ability to do high-throughput, multiplexed high-content screening with primary human liver cells. Traditionally, toxicity assays either sacrifice throughput or sacrifice biological relevance (using easy-to-grow immortalized cell lines instead of real human cells). We managed to combine throughput, physiological relevance, and multiplexing in one platform. The assays run in a high throughput format using automation, meaning thousands of compound-dose conditions can be tested in one experiment. We achieved this using pooled primary human hepatocytes, which are often fragile and expensive. By systemizing our automation and quality control processes, we were able to run over 120+ batches on the same donor pool with incredible reproducibility and consistency. We did this while integrating many readouts per well, whereas many existing toxicity assays only do a single readout. Our multiplexed approach provides far more data per experiment enabling us to measure 10-20 different toxicity phenotypes such as apoptosis, necrosis, mitochondrial fission, endoplasmic reticulum stress, stress granule formation, microtubules, and more all from a single well on a 384-well plate! The combination of scale, high content information, and data quality is exactly what is needed to train highly accurate AI models in biology. If you're interested, please explore the dataset in the comments below and let me know if you want to chat about the details!

Introducing the OpenHands CLI, a new coding CLI that: - Has top accuracy (similar to Claude Code) - Is completely open source, MIT licensed - Is model agnostic, use an API or bring your own - Is simple to install and run `pip install openhands-ai` and `openhands` (no Docker!)

🚨 New paper alert! Linear representation hypothesis (LRH) argues concepts are encoded as sparse sum of orthogonal directions, motivating interpretability tools like SAEs. But what if some concepts don’t fit that mold? Would SAEs capture them? 🤔 1/11

@hamiltonulmer as in code snippets/examples? Or actual if/then flow of when to do what?

My spicy take is that the best .cursor/rules is just well-structured code that a model can follow to do what you want, the way you want it

@ZanaBucinca OMG congrats!!!!

Thrilled to share that I’ve successfully defended my PhD dissertation and I will be joining MIT as an Assistant Professor starting Fall 2026, with a shared appointment between Sloan and EECS! I will be recruiting PhD students this upcoming cycle. Consider applying to MIT EECS!

Today we're launching DuckLake, an integrated data lake and catalog format powered by SQL. DuckLake unlocks next-generation data warehousing where compute is local, consistency central, and storage scales till infinity. ⁠DuckLake is an open standard and we've implemented it in the "ducklake" DuckDB extension. – Read the announcement blog post at duckdb.org/2025/05/27/duc… – Visit the new DuckLake website at ducklake.select

@zeddotdev @facebook found in extension docs, works great! github.com/zed-extensions…

@zeddotdev @facebook Is there a way to set which venv it should use?

We landed a Zed extension for `pyrefly`, @facebook's new Python language server written in Rust. "Type check over 1.85 million lines of code per second" and receive other goodies, like go-to-definition.

@umangsbhatt @Kings_College @Cambridge_Uni Congrats man, super exciting!

📣 Life update: I'm joining the University of Cambridge this fall as an Assistant Professor and as a Fellow in Computer Science at @Kings_College. Excited to (re)join the @Cambridge_Uni community!