Triomics

Triomics is a generative AI platform built for oncology workflows, transforming how cancer centres process and use unstructured patient data. Founded in 2021 by technical founders Sarim Khan and Hrituraj Singh, the company works with NCI-designated centres and leading academic and community oncology practices across the United States, and has raised $37 million to date.

Its core system, OncoLLM, is a specialised large language model that reasons across entire longitudinal patient records - hundreds of pages spanning years of care - rather than processing documents one at a time. It returns structured, cited answers with expert-level accuracy into existing EMR workflows. Around this sits a platform of AI agents that automate patient trial matching, visit preparation and pre-charting, and cancer registry curation across NAACCR, SEER, COC and QOPI measures.

The technical stack spans generative AI, large language models, natural language processing, clinical data extraction, unstructured data processing and agent-based systems. The platform processes more than 120,000 charts monthly and screens over 65,000 patients for clinical trials each month, and is used by four of the top ten U.S. cancer centres by U.S. News rankings.

The engineering problems centre on longitudinal patient record reasoning, medical-domain model accuracy, and integrating structured outputs into clinical systems. The stated aim of the work is to reduce the manual review burden carried by clinical teams.