Originals · Startup · Startups & Fintech
BioStack raises $9.5 Mn seed round led by Peak XV, DCVC and Afore to build training data for medical AI
The Y Combinator backed startup, which works with hospitals in India, the US, Brazil and Argentina, says it has access to more than 100 million de-identified patient records.
BioStack, a healthcare data startup backed by Y Combinator, has raised $9.5 million in a seed round. The Economic Times reported that Peak XV Partners, DCVC and Afore Capital led the round, and the company's own announcement on PR Newswire also names Emerson Collective, SV Angel, Verdict Capital, Streamlined Ventures and Y Combinator as investors. Entrackr and VCCircle carried the deal as well, with VCCircle putting the amount at about Rs 90.4 crore.
The company was founded by Sanat Mishra, its CEO, and Parth Patwa, its CTO, according to the announcement and Digital Health News. Peak XV's portfolio page lists it as founded in 2025 and based in the US, and the announcement carries a San Francisco dateline. India is one of its four operating markets alongside the US, Brazil and Argentina.
BioStack's pitch is that clinical data sits in pieces across hospitals and software systems. It builds workflows that pull together notes, lab results, imaging, prescriptions, treatment histories and outcomes into a single longitudinal record per patient, then strips identifying details and turns the result into datasets. The company says its network covers more than 100 million de-identified records, and that it has combined hundreds of thousands of X-rays, MRI and CT scans with pathology and other clinical records.
The buyers it is targeting are AI and biotech companies that need data to train and evaluate medical models, along with researchers working on disease modelling, biomarker discovery, patient stratification and treatment response. The announcement also says the datasets are meant to serve as reinforcement learning environments for medical AI agents.
The money will go into expanding in the US and other international markets, building out the technology platform, adding hospital partners and growing the engineering and research teams. For Indian readers, the model raises the obvious question of how patient data from Indian hospitals is consented and de-identified under the Digital Personal Data Protection Act, which neither the company's announcement nor the coverage so far addresses.
