Jul 2025 – Oct 2025 · 3 mos · Bengaluru, Karnataka, India

Jun 2023 – Feb 2025 · 1 yr 8 mos · Bengaluru, Karnataka, India · Hybrid

Jan 2022 – Jun 2023 · 1 yr 5 mos

• Migrated the Monolithic backend code base to Microservices architecture. This distributed system improved the user base, we could serve, ten times and made our downtime negligible.
• Implemented Validation Layer algorithm to improve accuracy for predicting bill date and due date of credit cards from 78% to 97%.
• Reduced batch processing time by 50- 60% using Spark for making card payments and collecting subscription fees.
• Implemented push based notifications to send progress report to our users and notify of upcoming payments using Clevertap.

Oct 2020 – Dec 2021 · 1 yr 2 mos

• Refactored Liability Manager, a Django based system using Postgres DB, by introducing tests and moving to Stateful architecture. Increasing our customer satisfaction and reducing tickets.
• Optimized most frequently visited apis of Debt Management System using parallel api calls and caching to reduce average api latency by 27%.
• Developed Credit Builder, to help users boost their credit scores by lending them small loans and ensuring timely repayment using security deposit.
• Designed and developed custom savings account feature - Asset Manager, using Firestore for FE integration.

Jul 2019 – Jun 2020 · 11 mos · Bengaluru Area, India

• Developed Lazy Match system to re-run the matching algorithm on unidentified video fingerprints with newer Ads identified in the last day and week. It runs during low traffic hours and has reduced daily unidentified clips by over 64%.
• Integrate C++ based Alphonso’s SDK with the vendor's AOSP TV platform. Also, customize and integrate android tv apps for TV OEMs.
• Developed a Javascript-based application that regularly analyzed the video clips database to detect and fix common data corruption issues.
• Created a JavaScript server to regularly scrape IMDB for top shows and used this data to refine
• the show matches for users.
• Worked on clustering IP addresses from the same household into one Devicegraph using Spark

Jan 2019 – Apr 2019 · 3 mos · Guwahati, Assam, India

• Using mechanism design theory, we design a truthful mechanism for solving the static load balancing problem in heterogeneous distributed systems. We prove that using the optimal allocation algorithm the output function admits a truthful payment scheme satisfying voluntary participation. Also we derived a protocol that implements our mechanism.

May 2018 – Jul 2018 · 2 mos · Bengaluru Area, India

• Worked to compare sequence prediction models CPT+, CPT and AKOM and modifying them to
• predict user behaviour based on user data obtained from smart home devices.