Alfredo Andere
@AlfredoAndere
Co-Founder and CEO @LatchBio — The Cloud for Biology
Biological data is growing exponentially. It’s difficult to reconcile and work with all of it. That’s why we’re building a Cloud for Biology. blog.latch.bio/the-cloud-for-…
Join our engineering team at Latch. Open technical problems at the intersection of compute + engineering biology. Fascinating customers. Real traction in a challenging market.
First time seeing this with a group of more than 3 people, we agreed to collaborate in Notion instead of Docs because it "copy-pastes correctly with chatGPT" and it "connects directly with Claude".
📜 ai doesn't run on just NVIDIA anymore - it’s running on many different chips, each with different quirks, tradeoffs, and scaling behavior. today we’re launching chipbenchmark.com - a new open-source platform to monitor the ai hardware situation.
It’s quite interesting how the technology used to spatially map nuclei at the nanoscale actually mirrors that of GPS, which works on an astronomical scale. In Slide-tag: A slide is coated with spatially barcoded oligos → Tissue is melted on top → UV light cleaves the oligos,…
Spatial biology is moving quickly and offers much to be excited about. But there is still a big learning curve. Especially with analysis. If you are a scientist new to spatial techniques, how do you actually analyze your data? A concrete tutorial from prep to insight follows:
Spatial biology is moving quickly and offers much to be excited about. But there is still a big learning curve. Especially with analysis. If you are a scientist new to spatial techniques, how do you actually analyze your data? A concrete tutorial from prep to insight follows:
I don’t understand how this continues to work
I was fired from Astronomer today. I ran our company’s events team. Turns out our CEO and Head of HR were having an affair and got caught at the Coldplay concert that I bought the company tickets to. They blamed me for getting caught and fired me on the spot. Taking some…
The beauty of Trekker is you can simply add it to the end of any single-cell experiment, and for ~$2K and a 1-hour add-on protocol, you get spatially resolved information. When the molecular biology side of a kit achieves that level of simplicity, the data analysis has a high…
looking forward to exciting times
Herdora (@herdora_ai) is the Cursor for CUDA. It automatically turns your PyTorch code into optimized GPU kernels so you don't have to write CUDA. Congrats on the launch, @technoabsurdist & @gpusteve! ycombinator.com/launches/NzG-h…
we built herdora because writing cuda sucks and hiring gpu engineers is impossible. we turn slow pytorch into fast gpu code. automatically. please reach out emilio [at] herdora [dot] com if you want faster/cheaper inference .
Herdora (@herdora_ai) is the Cursor for CUDA. It automatically turns your PyTorch code into optimized GPU kernels so you don't have to write CUDA. Congrats on the launch, @technoabsurdist & @gpusteve! ycombinator.com/launches/NzG-h…
Herdora (@herdora_ai) is the Cursor for CUDA. It automatically turns your PyTorch code into optimized GPU kernels so you don't have to write CUDA. Congrats on the launch, @technoabsurdist & @gpusteve! ycombinator.com/launches/NzG-h…
Just shipped Orbit. - Overlay running route & stats on photos - Track your PRs across any timeframe - Browse your run history like a commit log The app you use before you inevitably upload to Strava & Instagram. Details in 🧵
There is so much deep engineering work that is far upstream of drug discovery and in vivo systems. If you have a strong technical background you can have a very high impact by contributing to things like sequencing instruments, imaging, lab robotics and so on.
Imagine how things would reorganize around a whole transcriptome, epigenome + proteome spatial assay with intracellular resolution that can be prepped and read out in 10 minutes to your laptop from a $100 silicon wafer that fits in the palm of your hand These are fun things to…
Reminds me a lot of the recent wave of (very successful) systems companies that rewrote popular frameworks like Kafka to take advantage of modern storage devices. Emi is a super talented engineer + excited to see what he builds. Still so much software to write to close the gap…
📜 new blog post: amd’s mi300x gpu has huge potential for affordable, high-throughput llm inference - but it's currently underused due to software limitations. our initial optimizations already make it ~60% more cost-effective than nvidia's h100! (1/6) (🔗 links in final post)
📜 new blog post: amd’s mi300x gpu has huge potential for affordable, high-throughput llm inference - but it's currently underused due to software limitations. our initial optimizations already make it ~60% more cost-effective than nvidia's h100! (1/6) (🔗 links in final post)
China is surging ahead in NGS tech too, in both throughput & cost. 1. MGI's DNBSEQ-T20 → 21.6 Tb/day 2. GeneMind's Lightning → 9.3 Tb/day with Q40 3. Illumina NovaSeq X → 8 Tb/day Classic story of China excelling in scale & cost while US holding on to lead in innovation.
So our team is now about 60% AI at Team SaaStr. And it's ... a bit lonely. We're small. But every time someone has left, full time or part time, contractor or otherwise, we've now replaced them with AI. Mostly better. Not always, though. But almost out of necessity. But…
🔥 Introducing BindEnergyCraft (BECraft), the BindCraft pipeline you know and love, now enhanced with an energy-based loss to boost in silico binder success! Thrilled to present this as an oral at ICML @genbio_workshop next week 📄 Paper: arxiv.org/abs/2505.21241 🧵Thread⬇️