What the business is
An antibody-engineering platform pairing a proprietary wet lab with machine-learning models to design and test antibody candidates in days instead of months
The bet
A wet lab tightly looped with machine-learning models can engineer better antibodies faster than screening thousands of sequences ever could
Starting capital
$5.3M seed in 2019, $19M Series A in 2021 — $24.3M total raised
How it started
Bioinformatics veteran Mark DePristo and CSO Peyton Greenside founded BigHat in San Carlos, California in 2019, betting that combining computation with an in-house wet lab — not just software — was the way to out-engineer legacy antibody discovery.
What happened
BigHat raised a $5.3M seed in 2019 led by 8VC, built its wet-lab-plus-ML platform, and used it to engineer a neutralizing SARS-CoV-2 bispecific antibody as an early proof point. In February 2021 it closed an oversubscribed $19M Series A led by Andreessen Horowitz, with a16z general partner Vineeta Agarwala joining the board; prior investors 8VC, AME Cloud Ventures and Innovation Endeavors also returned.
What has to be true
Most antibody companies screen a large but fixed library of existing sequences, capturing only a tiny fraction of what's biologically possible.
Owning a wet lab meant BigHat controlled its own experimental data pipeline instead of depending on data other companies chose to publish or share.
Feeding fresh lab data straight back into the ML models in a tight loop let the platform target antibody designs that don't exist in any screening library.
An early proof point — a working SARS-CoV-2 bispecific antibody — gave investors evidence the loop worked before the Series A, not just a pitch.
What can be applied
When an entire industry screens the same pre-existing haystack, owning the data-generation loop yourself — not just the model on top of it — can be the real moat.
Aftermath
As of 2021-02-10, BigHat is using the new capital to hire antibody and protein engineers, AI/ML engineers and computational biologists, while continuing to build out its platform and push internal therapeutic programs toward human clinical trials over the next two to three years.
FOLLOW THE EVIDENCE