Chennai, Tamil Nadu, India
I build backend systems where applied ML lives inside production request paths. Currently an SDE on the KYC Hub team at CAMS, contributing to the cloud-native KYC platform used by AMCs and partner institutions across India. Day-to-day I work on Spring Boot (Java) REST APIs, partner onboarding (Apigee, JWT, RBAC), and third-party integrations like HyperVerge eSign — all running on GCP with AlloyDB. My path here started on a very different kind of factory floor. For my M.Tech thesis at IIT Kharagpur, I led an AI team at Ordnance Factory Badmal (Ministry of Defence), building a computer vision system to detect anomalies — cracks, piping, cavities — in X-ray images of critical manufacturing equipment. We annotated 25,000+ DICOM images, benchmarked YOLOv8 / YOLOv11 / RT-DETR, and reached ~85% precision on on-site validation. The client adopted it; the team is still extending the system. That combination — shipping CV in a regulated, high-stakes environment and building production backend systems in fintech — is what I keep doubling down on. On the side, I'm building VeriKYC (github.com/veriKYC/veriKYC) — a polyglot KYC platform (Spring Boot + FastAPI + React) that puts document detection, OCR, face match, and liveness checks behind a clean service boundary. It's my sandbox for the kind of systems I want to work on next. What I'm interested in: → Applied AI / ML engineering roles with real production deployment → Backend engineering for ML-heavy or document-intelligence systems → Fintech, KYC, identity, or industrial CV domains Open to remote / hybrid opportunities, particularly those based out of or friendly to Guwahati / North-East India. Tech I reach for: Java, Spring Boot, Python, FastAPI, PyTorch, YOLO family, PaddleOCR, OpenCV, PostgreSQL / AlloyDB, GCP, Apigee. Always open to a conversation.
Backend engineer on the KYC Hub team building Spring Boot (Java) REST APIs for investor-facing eKYC and client-facing Push APIs used by asset management companies and partner institutions — developing on the new GCP / AlloyDB cloud stack as part of CAMS's platform modernisation. Highlights: - Onboarded partners including PhonePe and Doctor Finance end-to-end — provisioned Apigee apps, configured Apigee-issued JWT auth, and designed RBAC across 10+ roles for client / partner / agent hierarchies powering the In-Person Verification (IPV) workflow. - Integrated HyperVerge APIs for Aadhaar-based eSign in the KYC journey — partner auth, retry handling, regulatory compliance. - Designed and implemented the API contract for an IPV change request enabling filtered retrieval and bulk export (CSV / Excel) of verification records. - Enterprise Delivery Hub (internal): owned deployment and end-to-end integration of an internal dashboard serving ~200 stakeholders across 12 roles — wired up frontend / backend / RBAC on GCP with AlloyDB, implemented Microsoft Entra SSO, and drove cross-team delivery with Infra, DevOps, and InfoSec.
Led an AI team to build a computer vision system for defect detection in industrial X-ray imagery at a Ministry of Defence ordnance factory. - Studied the factory's manual inspection workflow for anomalies (cracks, piping, cavities) and architected the CV replacement. - Built the DICOM pre-processing pipeline; used PaddleOCR to extract stamped number plates (serial no., batch, orientation) for per-unit traceability. - Added upstream classification models — one for outlier filtering, one for orientation correction — before feeding images to the anomaly detector. - Owned annotation and validation of 25,000+ DICOM images across multiple anomaly classes; set the labelling protocol used by the team. - Trained and benchmarked YOLOv8, YOLOv11, and RT-DETR — YOLO variants won on this dataset and were chosen for deployment. - Reached ~85% precision on on-site validation by end of tenure; client adoption led to a follow-on project for the company.
• Developed proof of concept projects using computer vision for real-time detection in an Agri IoT based solution. • Collaborated with the research and development team to create innovative solutions. • Utilized skills in computer vision to enhance project outcomes. • Contributed to the advancement of technology in the agriculture industry.