Capicú Foundation Model
Biomanufacturing AI with answers you can verify
1 follower
Biomanufacturing AI with answers you can verify
1 follower
Ask CFM what changed after a deviation, whether a batch stayed within approved limits, or which condition performed best. Built for biomanufacturing, the our model brings experiment plans, SOPs, batch records, deviations, and past decisions into one answer. It links each answer back to the source records, so scientists can decide what to do next without rebuilding the process history by hand—and help bring therapies to patients sooner.
Free Options
Launch Team
Render: Get up to $100K in startup creditsCloud infrastructure for founders who want to move fast and NOT break things
Promoted
Hi Accel AI members! I’m Sebastián, a computer scientist and bioengineer working on AI research and stem cell therapies. I recently found myself immersed in stem cell manufacturing through graduate school for the first time (ever!). It was hard, to say the least.
Simply put, in software I’m used to every change being tracked and commented. In the wet lab, a batch’s story is scattered across old instruments, handwritten diagrams, dense notes, and my lab mates “intuition”. Whenever you find yourself in the manufacturing line of a biotech company, big or small, you’ll encounter things like CAPA or root cause investigations. These can take 30 to 70 days to determine why a product didn’t meet the desired specs and often blame the person or operator on the line. Production stops, batches are lost, and when working with stem cell therapies, that batch may be a patient’s own cells.
So I, as a Spider-Man (Eve) & Iron Man (Jarvis) fan myself, what if the equipment processing this data could help us understand it? What if we could trace failures, characterize products without using them up, and plan the next experiment without starting from zero?
That question became our startup Capicú (https://capicu.ai). The vision began with AI inside lab instruments. It evolved into a practical first step: CFMs, models that read process history and points back to the evidence.
We’d appreciate people to try CFMs @ https://cfm.capicu.ai (there’s a free plan package too with a vision to release open-weights in the future!) Tell us where it helps and where it breaks. Your feedback, support, and shares genuinely mean a lot. <3