I’m a software engineer passionate about building reliable, scalable systems and turning ideas into real-world products. I primarily work with Java, Spring Boot, backend engineering, and distributed systems, while continuously exploring cloud, AI, and emerging technologies. I enjoy learning from the developer community, exchanging ideas, and understanding how technology can be used to solve meaningful problems. Always curious, always building, and always looking to learn from people smarter than me.
Software Engineer
Tech Mahindra
- • Develop Spring Boot microservices and RESTful APIs for business applications, handling validation, error
- handling, and pagination.
- • Work with PostgreSQL and Spring Data JPA to build database schemas, write queries, and ensure data integrity
- in transactions.
- • Integrate Apache Kafka to handle asynchronous events and messaging in distributed systems.
- • Debug application issues, test features across environments, and support releases.
- • Build Angular frontend components and their integrations with backend services.
- • Collaborated with frontend and QA teams to deliver features on sprint cycles, resolving production issues within
- SLAs.
LedgerX
• Built a financial ledger system using Spring Boot and PostgreSQL with immutable transaction records, ACID compliance, and crash recovery to prevent data loss. • Implemented pessimistic locking with deadlock prevention and API-level idempotency using database constraints to handle retries safely and prevent duplicate transactions. • Used async patterns, structured JSON logging with MDC context propagation, Docker containerization with health checks, and graceful shutdown to ensure in-flight transactions complete during deployment. • Built a reconciliation service to detect and resolve stuck transactions, managed schema migrations with Flyway, and implemented proper shutdown handling.
StoryPilot
Built a Python AI agent, exposed as an MCP server, that takes a Jira story ID, fetches the story through the Jira REST API, and generates proposed code changes for the developer to review as a diff. Implemented a RAG pipeline (chunking, embeddings, vector search) that retrieves only the code and docs relevant to each story, cutting LLM prompt tokens by ~40% compared with sending full context. Added a human-in-the-loop approval step: no change is applied until the developer approves, edits, or rejects the diff, which keeps output safe and auditable. Exposed the workflow through a FastAPI service, containerized it with Docker, and logged per-request token usage and latency to measure the efficiency gains.
Bachelor's of Engineering
Progressive Education Society's Modern College of Engineering, Pune · 9.10
Attendee · Oct 24, 2026