BioStack raises $9.5m to turn hospital records into medical AI data

BioStack raises $9.5m to turn hospital records into medical AI data

22 October 2026: BioStack, a San Francisco startup that turns scattered hospital records into datasets for training medical AI, says it has raised $9.5 million in seed funding. The company names Peak XV, Afore Capital, DCVC, Emerson Collective, SV Angel, Verdict Capital, Streamlined Ventures and Y Combinator as investors, and says the money will take it further into the US and international markets and grow its engineering and research teams.

Not yet independently verified. The company's announcement lists all investors together without naming a lead; Indian Startup News names Peak XV, DCVC and Afore Capital as leads. No valuation was disclosed. We will update this when it can be confirmed, and remove this note.

BioStack announced the round on its own site on 22 October. It took part in Y Combinator's Spring 2026 batch.

Following each patient across the record

Medical AI models are usually trained on data that arrives in pieces: a lab result in one system, an image in another, a discharge note in a third. BioStack's pitch is to link those pieces at patient level so that a model sees the whole course of care. Sanat Mishra, its co-founder and chief executive, put it this way: "Collecting every event linked to each patient enables a view of what happens at each stage of a patient's journey."

The company says it already works with hospital data across the United States, Brazil, India and Argentina. Building datasets in several countries matters for model makers who need patients of different ages, ethnicities and care systems, and who face separate consent and privacy rules in each.

The team and the investors

Mishra previously did cancer genomics research at Stanford, Yale, Carnegie Mellon and the Max Planck institutes, according to the company. His co-founder and chief technology officer, Parth Patwa, was a generative AI scientist at AWS. Indian Startup News reports that BioStack was founded in 2025 and has a team of 11.

Pushpak Kedia, a vice president at Peak XV, said in the company's announcement: "We believe BioStack has the potential to become an important data layer for the next generation of medical AI."

Where it sits among health data tools

The company says its footprint covers more than 100 million de-identified patient records, several hundred thousand X-rays, MRI and CT scans, and several million telehealth audio transcripts. Data on that scale raises questions buyers and partners will ask before any contract: how consent was gathered, which de-identification method was used, and which country's law applies to each dataset. The announcement does not go into them, and the company points to a trust centre for its security and privacy controls. Our catalogue lists 5 electronic medical records products, the systems such datasets are drawn from, and 1,080 artificial intelligence companies, among them the model makers that would buy them.

Company profile on TrustList: BioStack

Related on TrustList:

• Modal raises $750m Series D as AI compute run rate passes $1bn
• How to raise money for a startup, step by step
• Funding rounds explained: pre-seed, seed, Series A and beyond

Sources

• BioStack: announcements (seed round, 22 October 2026), 22 October 2026
• Indian Startup News: YC-backed BioStack raises $9.5 million to turn messy hospital records into training data for medical AI, 26 October 2026