Nov 2024 – Dec 2025 · 1 yr 1 mo · Bengaluru, Karnataka, India · On-site

• Built a robust settlement system integrated with the settlement bank, enabling automated reconciliation of 8+ lakh transactions and daily transfers of 100+ crore INR, while coordinating API-level integrations for seamless operations.
• Designed and implemented a closed-loop payments system for a hospital in collaboration with a partner bank, enabling card issuance, money loading, transactions, refunds, reversals, and card closure for seamless patient financial management digitally.

Apr 2022 – Aug 2024 · 2 yrs 4 mos · Bengaluru, Karnataka, India

• Built Search As You Type Service which enables search results across different verticals, catering to 10mn+ search queries and 150mn+ monthly active users across Paytm.
• Successfully extended the Search platform to accommodate Voice and Semantic search functionalities using GPT4 & Llama3 AI capabilities by writing effective prompts and then calling API-based/ES-based retrieval.
• Led a seamless migration of the Elasticsearch cluster from EC2 to a Kubernetes setup, ensuring zero downtime and maintaining 100% data consistency throughout the transition.
• Optimized Kafka cluster for higher consumption rates by implementing Snappy compression to reduce
• network overhead and enhance throughput. Led cost optimization to save ~40% over time. Reduced overprovisioned pods and also worked on security vulnerabilities.
• Contributed to implementing a localization feature, providing search results in users’ local/regional languages and enhancing the inclusivity and accessibility of the search platform for a diverse user base.
• Developed a Geohash-based location search feature, enabling search results relevant to the user’s location.

Sep 2019 – Mar 2022 · 2 yrs 6 mos · Bengaluru, Karnataka, India

• Paytm Mall
• Developed an A/B Framework and exposed it through the Django panel to test different features on some users. Able to track different experiments and associate CTR, impressions, etc to make a decision.
• Migrated Stats ES from v1.5 to v7.9 by rewriting the queries, making SLM policies, and storing data beyond 7 days in the S3 bucket leading to a significant reduction in infra cost.
• Implement auto-suggest to cater to prefix-based suggestions. Also, automated trending search flow to capture user data and show user-relevant results on a day-to-day basis.
• Paytm Travel
• Responsible for building inventory for bus providers in the IMS team. Golang, Kafka, and Cassandra were used primarily to implement the Producer-Consumer model.
• Integrated different provider APIs for seat layout, book, cancel, pre-cancel, etc in the IMS team.

Jul 2017 – Aug 2019 · 2 yrs 1 mo · Bengaluru Area, India