
Pathology AI Platform
AI pathology platform that delivers cancer detection and grading insights from WSI slides, with end‑to‑end product ownership and integrations to third‑party healthcare systems.
Pathology AI Platform
AI pathology platform that delivers cancer detection and grading insights from WSI slides, with end‑to‑end product ownership and integrations to third‑party healthcare systems.
Context
Oncologists needed faster, audit‑ready insights from whole‑slide images (WSI) while integrating with existing clinical systems and sustaining platform reliability across cloud and on‑prem environments
What We Engineered
A case/slide management workflow application with reporting and test automation.
AI analytics services (training, inference APIs, deployment) and annotation tooling for ground truth creation.
OpenAI‑powered chat for slide analysis, report summarization, and agentic flows; AWS & on‑prem Kubernetes deployments.
Intelligence Applied
Detection/segmentation/classification using Detectron2, ConvNeXt, ResNet+UNet; orchestration via LangChain/OpenAI for clinical‑grade summarization.
Impact Delivered
Accelerated slide triage and decision support for oncologists.
Unified workflow from annotation to model deployment and clinical reporting.
Enterprise readiness with multi‑env deployment (AWS + on‑prem K8s) and automated testing.
Highlights / Stack
React, Node; Detectron2, ConvNeXt, ResNet+UNet; LangChain, OpenAI; ArgoCD; Katalon/Studio; AWS & on‑prem Kubernetes.

