Trina Das

Biography
Trina Das builds data infrastructure enabling frontier AI systems to work reliably in precision medicine and diagnostics. She is the founder of Trinzz, a data platform designed to convert complex, multimodal medical data into high-fidelity, regulatory-ready datasets for training and evaluating diagnostic AI models. Combining expert-in-the-loop workflows, automation, and rigorous quality controls, Trinzz addresses data noise, bias, and distribution shift, empowering reliability in medical AI. Her work focuses on how real-world clinical data can be responsibly structured, annotated, and validated for AI models across diagnostic modalities including DICOM, NIfTI, ultrasound, digital pathology, and volumetric imaging. Trina’s background includes building and scaling large distributed systems across education and workforce technology, developing expertise in human-AI collaboration, quality standardization at scale, and incentive-aligned expert networks. Her contributions is recognized internationally, including Forbes 30Under30, Harvard alumni honoree and recognition at the White House by President Barack Obama as one of the most impactful emerging leaders under 25.

---

Talk
The Data Problem Behind Medical AI
Despite impressive model benchmarks, most medical AI systems break in real clinical settings and data edge cases. This session explores the hidden data pathologies behind failure, from annotation bias to distribution shift, and outlines how trustworthy data pipelines enable AI systems that meaningfully impact precision medicine and longevity.

AI and Data Sciences Showcase:
Trinzz Inc
Trinzz builds the foundational data infrastructure powering frontier-scale medical AI. It transforms complex, multimodal clinical data into reliable training and evaluation datasets using expert networks, automation, and clinical-grade quality assurance.