![](\"https:\/\/mms.businesswire.com\/media\/20260512133959\/en\/2802026\/5\/Copy_of_Knit_BrandMarkLogo.jpg\")
<\/a><\/a>\nMuch of AI in healthcare today relies on large language models primarily trained on text or published literature to support clinical operations and decision-making. This approach overlooks collective clinical intelligence: the patterns of real-world decision-making and coordination embedded in referral habits, scheduling practices, and the informal ways clinicians navigate complex institutions. This knowledge is what ultimately drives better outcomes, ensuring patients are routed to the right care, at the right time, with the right information.<\/p>
\nKnit Health is addressing this gap as it builds its LCBM <\/b>trained on <\/b>Truveta <\/b>EMR Data from over 130 million patients across 30 U.S. health systems. Using deep reinforcement learning, causal inference, and behavioral cloning, the model learns directly from clinician decision-making patterns and applies that insight across operational and care delivery workflows \u2013 enabling providers to deliver faster, more informed, patient-specific care pathways.<\/p>
\n\u201cMuch of what matters most in medicine isn\u2019t written in textbooks, it\u2019s learned through experience with time and navigating the healthcare system,\u201d said Jonathan Kolstad, Co-founder and CEO of Knit Health<\/b>. \u201cAcross millions of patient journeys, clinicians develop patterns for what to do next and when. Knit learns from those real-world decisions, transforming collective clinical experience into intelligence that improves how the system works.\u201d<\/p>
\nKnit Health\u2019s approach differs from traditional systems in several key ways:<\/p>
- \n
- \nBuilding to a behavioral world model:<\/b> Knit\u2019s LCBM learns from observed sequences of clinical decisions, enabling it to reflect how care actually unfolds in practice rather than producing probabilistic text.<\/li>\n
- \nHealth system-specific<\/b>: Knit fine-tunes to each health system\u2019s practice patterns, capacity constraints, and referral dynamics, allowing it to integrate seamlessly into existing operations and deliver measurable impact from the outset.<\/li>\n
- \nInfrastructure layer to drive health systems: <\/b>Knit sits beneath every routing decision, every discharge prediction, every care team allocation, every referral, and eventually every clinical workflow that touches a patient. This drives intelligence across the systems.<\/li>\n<\/ul>
\n\u201cKnit Health is creating a new approach to AI. Unlike traditional models, it learns and evolves from real human behavior and can be applied across complex systems,\u201d said Tripp Jones, General Partner at Uncork Capital<\/b>. \"This approach redefines how intelligence is captured and scaled, opening entirely new possibilities for AI-driven innovation in healthcare.\u201d<\/p>
\n\u201cThe hardest challenge in healthcare isn\u2019t knowing what good care looks like, it\u2019s delivering it consistently for every patient,\u201d said Navid Farzad, Managing Partner at Frist Cressey Ventures<\/b>. \u201cKnit Health\u2019s model embeds the best clinical intelligence directly into the workflow, helping clinicians make better decisions faster and more consistently. At scale, this will improve patient outcomes and transform clinical operations across health systems.\u201d<\/p>
\nKnit Health is built with full HIPAA compliance \u2013 rigorous governance, bias testing, and continuous monitoring to ensure guidance that patients and providers can trust. The company is partnering with health systems to deploy its initial models for triage, patient flow, and quality improvement. Through these collaborations, Knit aims to become the irreplaceable infrastructure layer that every clinical decision runs on.<\/p>
- \nHealth system-specific<\/b>: Knit fine-tunes to each health system\u2019s practice patterns, capacity constraints, and referral dynamics, allowing it to integrate seamlessly into existing operations and deliver measurable impact from the outset.<\/li>\n