Getting the First 69% For Free With Decision Models
I zero-shot labeled much of a bioreactor dataset using an open-weight decision model, local compute on my laptop, and prompts so basic that they're a little silly.
I zero-shot labeled much of a bioreactor dataset using an open-weight decision model, local compute on my laptop, and prompts so basic that they're a little silly.
👉This is a light treatment of LLM usage in biotech automation. I'm still looking for the killer application that LLM technology solves. It's not direct control. Of course LLM's have been the hot topic in computing for some time, and, as with any buzzy
🔁This post was originally made as a lab note for BioArtBot's Experiment.com project to build a place-based microbial palette for bio-art. Our San Francisco Bay Area microbial palette is in progress, and I wanted to share with you an interesting microbial story that has come
Automation integration is the work of connecting different automation tools together to get something specific done. Recently I reviewed of three recent proposals to make automation faster and easier, and presented them at Open Automation Club: * BioPhorum's A framework for the requirements for modular process equipment: BioPhorum Plug
🔁This post was originally made as a lab note for BioArtBot's Experiment.com project to build a place-based microbial palette for bio-art. I wanted to share some of the strains we've been researching. I live in the San Francisco Bay Area, and it'
👉This post is a technical proposal describing a framework for building automated systems for science. The framework executes scientific experiments in specific, high-context environments, while keeping the scientific ideas in those experiments in a general, sharable form. Science is hard - we should automate it. But automation integration takes forever.