BigHat Biosciences

Company Overview

US biotech company BigHat Biosciences builds protein therapeutics with machine learning, and its argument is less about the models than about the data that feeds them. The company generates its own biological measurements at high throughput and uses them to train the AI that designs the next round of molecules. For six years that was a platform story. It is now a clinical one: BHB810, an antibody-drug conjugate aimed at gastrointestinal cancers, has dosed its first patient, and the company calls it one of the first AI-driven biologics to reach human trials. In September 2026 BigHat closed a $75 million Series C, taking total funding to $223 million.

Headquarters and Global Presence

BigHat is headquartered in San Mateo, California, on the San Francisco Peninsula. That places it between the Bay Area's software talent and its biotech cluster, which is the combination a company pairing frontier AI with wet-lab automation needs to hire from.

Founding and History

BigHat was founded in 2019 by Peyton Greenside and Mark DePristo, formerly head of genomics at Google AI. Its July 2022 Series B, $75 million led by Section 32, brought in Amgen Ventures and Bristol Myers Squibb. It then called itself an antibody company; by 2026 it had widened that to protein therapeutics and put its first molecule into patients.

Therapy Areas and Focus

Oncology leads. Both disclosed programs with a named target are aimed at solid tumors: BHB810 at CDH17-positive cancers of the upper and lower gastrointestinal tract, and BHB299 at tumors expressing CEACAM6. A third T-cell engager, BHB-AV, is listed against an undisclosed target in inflammation and immunology, and a fourth, BHB-VH, in oncology. The concentration on gastrointestinal cancers reflects a target choice rather than a disease specialism: CDH17 is expressed across colorectal cancers and in most gastric cancers, so one antigen opens several indications.

Technology Platforms and Modalities

The platform, which BigHat introduced as Milliner, couples machine learning with an automated, high-throughput wet lab. Designs proposed by the models are made and measured, and the results go straight back into training, so each cycle improves the next. The company now frames this as an integrated learning loop producing gold-standard biological data at scale, and its Series C release describes an agentic data platform for rapid protein design. The point of the loop is to optimize several properties at once, potency, selectivity, developability, rather than one at a time. The modalities in the pipeline show the range: an antibody-drug conjugate, avidity-driven T-cell engagers and a T-cell engager with an added payload.

Key Pipeline and Programs

BHB810 is a CDH17-directed antibody-drug conjugate and the lead asset. It is in a Phase I/II study in advanced gastric and gastroesophageal junction adenocarcinoma that began in July 2026 and is recruiting, with colorectal and pancreatic cancers and other CDH17-positive solid tumors also in scope. The first patient has been dosed. BHB299 is an avidity-driven T-cell engager built to bind CEACAM6 and CD3 and to target CEACAM6-expressing solid tumors selectively, using avidity rather than raw affinity to discriminate. It is completing preclinical work, with human trials planned for 2027. Behind those two sit BHB-AV, a T-cell engager for inflammatory and immune disease, and BHB-VH, a T-cell engager-plus construct in oncology, both against undisclosed targets. Everything clinical rests on BHB810, which makes its first dose-escalation data the event that will test whether AI-designed biologics behave differently in people.

Recent Developments

On 24 September 2026 BigHat closed a $75 million Series C co-led by DFJ Growth and Premji Invest. Catalio Capital Management, LG Technology Ventures and Sigmas Group joined, alongside returning investors including Amgen Ventures, Andreessen Horowitz, Eli Lilly and Company and the Merck Global Health Innovation Fund. The money goes to the data platform and to clinical development of BHB810 and BHB299. Earlier in the year the company strengthened its research leadership, appointing a Roche veteran as chief scientific officer in June 2026. The investor list says something on its own: three large drugmakers are shareholders as well as partners.

Key Personnel

Peyton Greenside, co-founder and chief executive, frames the business as combining frontier AI with the autonomous generation of biological data. Stefan Weigand, chief scientific officer since June 2026, was head of large molecule research at Roche and brings big-pharma biologics experience to a company founded by computational scientists. John Corbin, chief development officer, was chief technology officer at Tizona Therapeutics and carries BHB810 through the clinic.

Strategic Partnerships

Partnering has funded the platform while the pipeline matured. BigHat completed a collaboration with Johnson & Johnson, has worked on three projects with Merck & Co and has two therapeutic collaborations with Eli Lilly, extended through Lilly TuneLab. Amgen, Lilly and Merck also hold equity, so the partners are betting on the approach, not just buying services.

FAQ Section