There is a quiet, exhausting reality inside every major cancer center. Each month, highly trained clinical staff are forced to manually review tens of thousands of patient records to match them with life-saving clinical trials. This means sifting through hundreds of pages of unstructured data per patient — pathology reports, scanned faxes, genomic panels, and years of clinical notes — just to evaluate eligibility against complex criteria. We are asking brilliant oncology teams to act as human data scrapers. When a system runs on sheer manual effort, the result is tragic but inevitable: clinical coordinators burn out, and eligible patients miss out on treatments that could save their lives.
It takes builders with deep engineering rigor and an uncompromising focus on clinical safety to look at this overwhelming bottleneck and build the infrastructure to solve it at scale. That is exactly what Sarim Khan and Hrituraj Singh are doing with Triomics.
Founded in 2021, Sarim and Hrituraj are building a deeply intelligent, autonomous AI platform purpose-built for the immense complexity of modern oncology.
Instead of relying on a single, generic language model, Triomics deploys OncoLLM — a constellation of specialized models that work agentically to reason at the patient level. Paired with OncoIndexer, which ingests and normalizes every messy data type from HL7 feeds to scanned PDFs, the platform reads the full longitudinal patient chart. It seamlessly integrates directly into major EHRs like Epic and OncoEMR, automatically processing medical records at scale and in real time.
The true moat here is uncompromising clinical explainability and workflow integration. Triomics doesn’t just generate a guess; every output is source-backed, cited, and instantly verifiable inside the clinician’s existing workflow. By automating the agonizing work of pre-visit chart reviews and clinical trial matching, Triomics enables oncologists and their teams to evaluate complete records against hundreds of ongoing trials in under a minute. It shifts the paradigm from hunting for data to proactively acting on what matters most.
The market? Leading health systems, NCI-designated cancer centers, and community oncology practices across the US. Triomics is already trusted by 4 of the top 10 cancer hospitals in the country, processing millions of clinical documents a month and driving massive lifts in trial accruals.
Seeing founders head-down, quietly architecting the heavy-duty infrastructure that fundamentally accelerates cancer research and care, is profoundly inspiring. They are ensuring that every patient has a fair shot at the right trial, giving clinicians their time and their capacity to care back.
Every founder dreams of reaching orbit. Sometimes, all it takes is the right launch window. If you are building an AI-first healthcare startup and looking to experience the US market firsthand, today is the final day to apply for Vertical Velocity Cohort 2. Applications close tonight: https://aiboomi.org/programs/vertical-velocity-26/
Let’s celebrate the builders.
w/
#HealthTech #Oncology #ProductNation
