Layer Health cofounders Luke Murray, Monica Agrawal, Divya Gopinath, David Sontag, and Steven Horng
Artificial intelligence (AI) is enjoying its moment as the hottest area of venture investment, with more than $100 billion flowing into the sector last year. In healthcare, AI accounted for 30% of all venture funding in 2024 – and data shows 2025 is off to a strong start.
That momentum continues today with the announcement of a $21 million Series A by Layer Health, an ambitious healthcare AI startup aiming to tackle some of the sector’s thorniest issues and overcome the industry’s biggest barriers to growth.
The round was led by Define Ventures, with participation from Flare Capital Partners, GV and MultiCare Capital Partners. They join a cap table that already includes General Catalyst and Inception Health, which suggests credibility in the company’s approach.
Layer Health is applying large language models (LLMs) to perform data abstraction for medical chart reviews. Seemingly mundane and esoteric to the outsider, chart reviews are a foundational task that underpins a wide range of clinical and administrative workflows within health systems (and for other ecosystem partners). They can entail combing through vast volumes of some of the most fragmented and complex data in any industry – medical records – to answer highly specific, context-rich questions.
Whether used to support clinical decision-making at the point of care or for administrative functions like clinical documentation improvement (CDI), chart review remains labor-intensive and highly technical. Depending on the use case, it can require scouring both structured and unstructured data – visit records, progress notes, imaging reports, lab results – and interpreting it with clinician-level understanding. At scale, this process becomes expensive and time-consuming, especially since it’s currently often performed manually by highly trained professionals.
These characteristics make chart review particularly well-suited for AI. LLMs excel at processing, summarizing and interpreting unstructured data with speed and precision. While there have been issues with LLMs “hallucinating” at times, Layer Health contends that its models, which are trained on longitudinal data, can support its outputs with cited evidence, helping end users trust and verify the information presented.
Still, deploying LLMs in real world healthcare settings – especially across disparate clinical environments – is no easy feat. Layer Health, which emphasizes the flexibility of its core AI platform and its ability to mitigate the hallucination problem, is navigating a complex and competitive market. Yet its founding team’s deep experience and system-aware approach to the unique challenges of healthcare organizations could help differentiate it.
Peeling Back a Layer
While most high schoolers in the late 1990s were focused on malls, Nintendo 64 consoles or chatting on their Nokia phones, Layer Health co-founders David Sontag and Steven Horng were already discussing how they might one day make an impact on the world. Both were drawn to computer science and shared a strong entrepreneurial drive.
Like many teenage friends, they eventually pursued separate paths. Sontag earned a Ph.D. in computer science and held faculty positions at New York University and the Massachusetts Institute of Technology. Horng went on to become a physician, earning additional degrees in computer science and biomedical informatics. He currently serves as an attending emergency physician at Beth Israel Deaconess Medical Center, where he also leads machine-learning initiatives.
Both had promising, independent careers, but their desire to collaborate eventually brought them back together. Horng’s day-to-day experiences in the ER gave him first-hand insight into the complexity and inefficiency of healthcare workflows and data systems.
Starting in the early 2010s, the pair began building test applications – often with Sontag’s students – within Beth Israel’s (at the time) homegrown EHR. Over time, they explored a range of AI use cases for both clinical and administrative teams, iterating across many early models.
“We originally deployed an algorithm for detecting sepsis but quickly detected that was not where we were going to have a big impact,” said Horng. “After making that discovery early, we pivoted to clinical workflow.”
As LLMs began emerging as a transformative force in AI, the groundwork for Layer Health started to crystalize. One of the first widely cited papers on the use of LLMs in healthcare was co-authored by Sontag and another eventual co-founder, Monica Agrawal, a former MIT student who now is also a professor at Duke.
By 2022, the collective experiences of Sontag, Horng, Agrawal and two additional former MIT students, Luke Murray (a software engineer from Google and SpaceX) and Divya Gopinath (a founding engineer at Snowflake-acquired TruEra), led to the formal founding of Layer Health.
Layering In The Name
While medical chart data abstraction is at the heart of Layer Health’s AI platform, its modular architecture is key to the company’s strategy, according to Sontag and Horng. Each module supports a specific function but also contributes to and builds upon the others, enabling the system to learn and improve across use cases.
The company’s initial focus is a module that supports clinical registry reporting, which are used to track outcomes over time and support research, quality improvement and public health. The module has been deployed already at Froedtert & the Medical College of Wisconsin health network, where it was used to abstract data for quality reporting. According to Layer Health, its AI reduced the required time by “more than 65%.”
From there, Layer plans to validate one of its next modules: real-time clinical decision support at the point of care.
“The same chart review problem we’re solving with our clinical registry module is faced by clinicians at the point of care,” said Sontag. “For example, one of our next modules will focus on real-time clinical decision support to help automate clinical care pathways, leading to more reliable, high-quality care. This will not only improve patient outcomes, but will also naturally lead to more timely and accurate revenue capture, quality improvement and research.”
Additional modules under development aim to support hospital operations and revenue cycle management by enhancing CDI and medical coding processes. The broader vision is to offer an enterprise-level solution – a foundational AI “layer,” as the name implies – that spans departments and delivers cumulative ROI over time.
Chart review isn’t just essential for providers. Life sciences companies and clinical research organizations also rely on it to answer highly specific, nuanced questions, especially when evaluating patients for clinical trial eligibility. Manually reviewing charts to assess thousands of patients against inclusion and exclusion criteria is slow and costly, making it another ripe area for automation.
Layer Health recently signed a multi-year agreement with the American Cancer Society (ACS), which will use its platform to extract clinical data from thousands of patient records tied to research studies, including the Cancer Prevention Study-3. The deal followed a successful pilot in which the AI accurately abstracted real-world data in a fraction of the time.
Layering On Competition and Bespoke Problems
Despite promising early traction, Layer Health faces a significant battle in a competitive market within an industry that’s notoriously difficult to scale. Health systems often struggle with people- and process-related challenges that can’t be solved by technology alone. Even within the same organization, different departments may have unique configurations, workflows and legacy systems that complicate implementation.
The idea of a transferable, enterprise-wide AI solution is appealing, but in practice, significant barriers remain. Layer Health acknowledges these complexities and believes its platform is designed to meet them head-on.
“While many of healthcare’s challenges are universal, some are uniquely local. Our enterprise platform also makes it possible for hospitals to easily configure, evaluate and deploy AI for chart review for their specific, local problems. It directly integrates with a hospital’s electronic medical record and existing business intelligence platforms, easily extending a hospital’s existing workforce to use AI chart review in a no-code / low-code way. The self-service SaaS platform is already in use by our life science customers,” said Sontag.
Investors share this belief. Lynne Chou O’Keefe, founder and managing partner at Define Ventures, sees Layer’s architecture as a key differentiator.
Lynne Chou O’Keefe is Founder and Managing Partner of Define Ventures
“Layer Health is designed to be a foundational AI platform, rather than a single-use AI tool. Many AI solutions in healthcare are highly specific to a single workflow or require extensive customization for each customer,” O’Keefe said. “In contrast, Layer Health has built a generalizable LLM-based system that can interpret complex clinical data across multiple use cases. Its AI reasons across an entire patient chart, allowing health systems to derive clinician-level insights with minimal configuration. This ability to scale across different health system environments without excessive customization is a key differentiator.”
Define Ventures, which previously announced $460 million across two new funds, saw Layer as a natural fit for its investment thesis.
“We believe the most successful AI companies will be those that solve deep, system-wide inefficiencies rather than offering surface-level automation. Layer Health embodies this thesis by addressing the immense problem of clinical data abstraction and chart review, a historically manual, error-prone, and resource-intensive process. By creating a generalizable AI infrastructure for clinical inference, Layer has the potential to become the foundational AI layer for healthcare organizations, making it a natural fit for our investment approach,” explained O’Keefe.
Flare Capital Partners also sees value in Layer’s low-friction deployment model and revenue-generating potential for health systems operating on tight margins.
Parth Desai is Partner at Flare Capital
“Layer Health’s AI platform uncovers powerful revenue-generating insights for health systems, through a unique ability to unify clinical chart data with outcomes. Powered by breakthroughs in AI, Layer Health can also deliver these insights in real-time, with minimal integration and at a fraction of current costs,” said Flare Capital Partners Partner Parth Desai. “This has made David and team a foundational and trusted partner to all healthcare organizations deploying AI.”
The Final Layer?
Layer Health’s goal to become the connective AI tissue across clinical, operational and research domains is ambitious. With early traction in clinical registry reporting and expanding partnerships across the provider and life sciences sectors, the company is positioning itself as more than a single-use solution. However, the path to widespread adoption in healthcare will demand not just technical strength, but also adaptability to deeply rooted workflows and fragmented infrastructure.
Backed by $21 million in fresh capital and investors betting on foundational impact, Layer Health now faces its next challenge: demonstrating that its platform can scale, deliver meaningful ROI and adapt to healthcare’s complex realities. If successful, the company may not only set itself apart in a crowded AI landscape—it could help define how large language models are integrated into the future of healthcare.
