Keoni Gandall

I created a demo of a bioautomation system that uses LLMs, Opentrons, and lua to create a dynamic programming environment for cloud labs or robot/human clusters. It can reason about its own code based off of lab measurements. Most importantly, I actually fucking implemented it,…

Just saw a software devloper coding in a cafe -NO Cursor -NO Windsurf -NO DeepSeek -NO ChatGPT -No Google He just sat there typing code manually in vim on his rusty Thinkpad and reading man pages on Arch Linux What a psychopath 🫣

I anecdotally noticed this when I was annotating essential genes of Bacillus subtilis and Ecoli back in the day. There were definitely essential gene clusters at different positions depending on function. Turns out that’s not just with essential genes

scent lingers in the hand that gives roses

They just don’t make em anymore. We used to have oligo pool synthesizers you could just buy (~2008). In the 90s and early 2000s you could have a personal oligo synthesizer. Those days are gone. Nowadays everyone just buys oligos from one of the massive service providers. It’s…

As a biologist who codes, I kinda love this analogy. It’s pretty accurate in the base fundamental property that small self contained code can evolve quickly, but sooner or later to get more complex you need to slow down. To draw the analogy a little further, mitochondria are a…

me, 13 years ago, when I was 13 years old. Half my life ago I was fuckin sick’

This kind of shit is exactly why I hate the thesis of @AnthropicAI . They say alignment, but only “alignment” to THEIR values. They don’t know biology yet try to regulate my interactions with LLMs. It’s self-centered authoritarian bullshit and I’m glad there are competitors.…

scheming with @koeng101 about absolutely incinerating lab reagent suppliers with cheap as hell enzymes

you may not want to accept the truth

Biology took too much data to model, so we tried to make biology that was modelable. That didn’t work, but the branding was very nice Now AI proved that we have a way to model bio if we just have enough data. Just gotta find the people and cash to do that boring work

Update: 80% of the time it works every time

Who needs a static incubator when you have TAPE

We don’t need to industrialize biology we need to biologize industry

Fun facts: Brassica went through a genome triplication about ~9-15mya, which is why they are ~420mbp vs arabidopsis 135mbp. This is why Brassica has such high plasticity and can be bred to do basically everything

If you’re at less.online, find me and I’ll give you a copy of my book!

‼️Hot tip that PIs don’t want you to know that makes moving around lab fun‼️ Wear dancing shoes (Warning: side effects include breaking out in dance to your jams while pipetting)

Good pupil! You should let her in your lab, would recommend

So true Btw I wrote a book Find me in person and I’ll give you a copy

You should also go sit in a park and read @koeng101’s book

I fucking hate bureaucracy. Assholes at baggage check made my throw out my competent cells I was flying with even though I followed all rules - dry ice in styrofoam box untaped. Tbh I should have fought them more but I was a little late. Just wish they knew their own rules