Alterra Labs raises $12.5M seed for medical AI

Alterra Labs announced a $12.5 million seed round on November 10, 2026, as it pushes deeper into simulated medical environments for training and evaluating AI models. The New York startup said the round was led by Oak HC/FT, with participation from Flare Capital Partners, Company Ventures and Tau Ventures.

Alterra’s core product is a set of reinforcement-learning environments and evaluations designed for clinical and scientific work. The company says it tests AI models on medical tasks, identifies where they fail and then converts those failures into training exercises. In its own announcement, Alterra said it is building an AI safety and alignment research lab for healthcare and life sciences, while Fortune reported that founder and CEO Akash Kapoor said current models can operate software but still lack clinical judgment and uncertainty handling.

What does Alterra Labs actually build?

Alterra Labs builds simulated healthcare environments that let AI systems practice tasks before they are used in real-world settings. According to Oak HC/FT, those environments simulate electronic health records, PACS, payer portals, clinical-trial software, consumer applications and scientific research tools. Flare Capital Partners said the company also reproduces multiple EMRs, imaging systems, a payer portal and voice workflows.

The startup’s workflow combines simulation, evaluation and post-training data. Alterra says clinicians help design simulated medical tasks and assess model decisions, then those assessments are converted into scoring criteria and reinforcement-learning rewards. Flare said each task includes a clinician-authored rubric, a verifier and a failure analysis, while Oak HC/FT said the company’s training signals are designed to assess not just answers but whether model reasoning and decisions meet professional standards.

Why are investors backing a healthcare-first AI environment company?

Investors say Alterra is filling a specific gap in AI training infrastructure for medicine and life sciences. Oak HC/FT said most reinforcement-learning environment and evaluation companies are horizontal, while Alterra is healthcare-first. Flare Capital framed the company as building simulation environments, graded evaluations, expert clinical judgment and post-training data for healthcare AI.

The investors also highlighted the scale of demand for training infrastructure. Flare said frontier labs improve models through post-training in interactive environments using verifiers to score whether tasks were completed appropriately, and added that labs have reportedly considered spending more than $1 billion on environments and related infrastructure. Oak HC/FT said AI labs already spend about $7 billion annually on training data and that spending is on pace to approach $100 billion by 2030.

Oak HC/FT partner Vig Chandramouli led the investment, according to Fortune. The firm said it worked with Alterra beginning in summer 2026 on a new name, visual identity and website, resulting in the Alterra Labs rebrand.

How will Alterra use the new capital?

Alterra and its backers said the seed funding will support expansion of the company’s environments, expert network and commercial work. Oak HC/FT said the money will help accelerate research, scale the in-house team, expand the expert-in-residence network and deepen partnerships with organizations developing and deploying advanced AI. Flare said the financing will fund expansion across Alterra’s three segments: environments, clinical expert network and commercial team.

The company’s in-house team includes more than 40 practicing clinicians, along with scientists and machine-learning researchers, according to both Oak HC/FT and Flare. Oak HC/FT said those clinicians are U.S.-trained and licensed physicians, while Flare said multiple attending physicians review every rubric and that clinicians and researchers jointly review model trajectories for final failure analyses.

Fortune reported that Alterra works with several frontier labs, hospitals, medical societies and applied AI companies, and Kapoor said frontier AI labs are the startup’s fastest-growing customer group. He also told Fortune that the company earns revenue from training environments and post-training services. Flare said Alterra has delivered calibrated sample sets to most leading frontier labs and converted some early engagements into paid partnerships with a handful of them.

What happened before the seed round?

Alterra started in 2025 while Kapoor was attending Columbia Medical School, according to Fortune. The company was originally called ChartR and initially worked with healthcare providers to study how frontier models behaved in clinical settings before shifting to work directly with model developers.

That pivot came after Alterra observed weaknesses in long-horizon, information-heavy and sequential decision tasks, Fortune reported. Kapoor, who studied machine learning, did healthcare-focused machine learning research at the NIH and completed three years at Columbia Medical School, left medical school in his final year to work on Alterra full time.

Flare said its first meeting with Kapoor included a live simulation of a health system environment with an EHR, coding workbench and payer portal. The firm said Alterra can deliver production-grade environments within weeks of a signed contract, and that it has established partnerships with hospitals, specialty societies and other healthcare organizations.

Together, those details suggest Alterra is positioning itself as infrastructure for a growing segment of healthcare AI: not just model building, but training, evaluation and post-training support that attempts to measure whether systems can reason, escalate uncertainty and behave like clinicians would expect before deployment.