Backend Engineer at Cognizant · ML Systems Builder
I architect production-grade backend systems in Java & Spring Boot, and train deep learning models that ship to real users. Currently building scalable enterprise systems at Cognizant while pushing ML models into production on Hugging Face.
Every project ships to real users. No demos, no toy implementations — real engineering with measurable outcomes.
Full-stack enterprise platform with multi-module RESTful backend spanning policy lifecycle, secure payment workflows, and role-based portals. Secured with BCrypt hashing, AuditService logging every transaction, and deployed live via multi-stage Docker on Render cloud.
EfficientNetB3 model trained on 120K+ images (CIFAKE dataset) achieving 93.55% classification accuracy. Features Grad-CAM explainability via GradientTape with heatmap overlays. Deployed live on Hugging Face Spaces with thread-safe prediction and cold-start auto-download.
Benchmarked KNN, Random Forest, Logistic Regression & Decision Tree on 284,807 real transactions. KNN achieved 99.56% accuracy and 92.10% fraud sensitivity, selected via sensitivity-focused model evaluation to minimize missed fraud.
Integrated 9 government road-safety datasets spanning 37 states; built a 4-sheet Qlik Sense dashboard (Overview, Demographics, Safety Deep-Dive, Insights) enabling drill-down accident hotspot analysis for data-driven decisions.
Own and engineer production backend services across the full Java/Spring Boot stack. Operate in live Agile sprints, diagnosing production incidents, optimizing JPA queries, and shipping clean, maintainable code across Controller → Service → Repository layers.
Sole owner of the Payment Management Module for a live enterprise Auto Insurance System. Designed payment gateway simulation, retry logic, and persistent audit trails, achieving 95%+ code coverage through rigorous TDD with JUnit & Mockito.
Analyzed real government road-safety datasets across 37 Indian states using Qlik Sense. Delivered interactive KPI dashboards that surfaced actionable accident hotspots and demographic risk patterns for data-driven policy decisions.
Graduated with strong foundations in algorithms, software engineering, databases, and ML. Completed multiple live projects including AI Image Classification and Enterprise Insurance System as final-year major projects.
Looking for a traditional format to share with your team? Download the comprehensive 1-page PDF detailing my full-stack engineering workflow, API design, and ML pipeline implementations.
Open to backend engineering, ML engineering, and full-stack roles. If you're building something interesting, I'd love to talk.