Senior Software Engineer at CARPL - Radiology AI Platform

Designed and deployed an autonomous model monitoring system that detects drift and reliability issues in real time, reducing false negatives by 30% and improving AI stability across 10+ hospital sites. Led on-prem training platform with automated validation flows for data annotation, training in-house models, and scalable deployment for healthcare customers, reducing dev cycle from 4 weeks to 5 days. Architected scalable inference pipeline processing over 1M patient images and AI reports with load balancing and asynchronous task orchestration. Led a team of 4 engineers across monitoring, training, and deployment initiatives delivering high-impact AI solutions for healthcare customers.

Software Engineer & Technical Pre-Sales at CARPL - Radiology AI Platform

Built a multi-cloud storage microservice in GoLang, integrating AWS S3, Azure Blob, and GCP storage under a unified API; scalable to 1M+ requests/day with high reliability. Led the development and deployment of a scalable edge-based AI inference platform for early detection of tuberculosis in India, impacting 50K+ patients daily across rural clinics. Developed and fine-tuned a deep learning model leveraging CheXpert dataset for chest X-ray analysis, achieving precise localization of abnormalities using CNN architectures and attention mechanisms. Integrated HL7 FHIR-based APIs enabling standardized interoperability across health care systems and third-party applications. Redesigned the DICOM processing pipeline, reducing average inference latency by 40% and lowering cloud compute costs by ~25% at scale.