Jul 2024 – Jan 2026 · 1 yr 6 mos · Bengaluru, Karnataka, India · On-site

• I lead the backend architecture for Navi's collections orchestration platform—a system that processes over 1 million case assignments daily with industry-leading reliability (<40 errors/week across 15+ services).
• Key Contributions:
• System Design & Architecture
• Designed and scaled a multi-module collections platform with CQRS architecture—command side handles lifecycle operations while the query side leverages Elasticsearch with nested documents for sub-second search performance.
• Built a domain event processor using JOLT transformations and dynamic code loading, enabling clean separation between services and real-time event routing.
• Performance Optimization
• Revamped the auto-allocation engine with EMI/capacity-based models and introduced a shadow testing framework—cut allocation time by 60% (from 45 minutes to 16 minutes) and enabled 1,000+ dry runs per month for safe experimentation.
• Migrated segmentation from batch processing (6-8 hour stale data) to CDC-driven real-time pipelines, achieving 93% faster execution (30 min → <2 min).
• Platform Engineering
• Created centralized reactive Kotlin client libraries (WebFlux + Resilience4j) used across 15+ microservices, eliminating 3,800+ lines of duplicated code and standardizing resilience patterns.
• Developed a Field Allocation Self-Serve system enabling operations teams to configure allocation rules without engineering support—saving 21+ person-days monthly.

Jul 2023 – Jul 2024 · 1 yr · Bengaluru, Karnataka, India · On-site

• My first year at Navi focused on building core microservices and establishing engineering best practices for the collections platform.
• Highlights:
• Built microservices powering case lifecycle management, intelligent allocation, and customer segmentation using Spring Boot, Kotlin, and Kafka.
• Designed a modular activity execution framework with audit trails, trigger handling, and configurable business strategies—accelerating feature iteration for product teams.
• Migrated legacy re-allocation and tele auto-allocation workflows from S3-dependent batch jobs to a resilient real-time API platform, improving latency and fault tolerance.
• Authored 50+ component and integration tests for orchestration pipelines, significantly improving deployment confidence and reducing production incidents.
• Rewrote a Java distance-matrix routing service to Kotlin coroutines with intelligent response caching (GraphHopper + Google Distance Matrix API), reducing API costs while improving accuracy.
• Led a data cleanup initiative that deprecated 800GB of redundant OLTP data and optimized Spark geospatial jobs, saving $1,100/month in infrastructure costs.

Jun 2022 – Aug 2022 · 2 mos · Banglore · Remote

• Contributed to the core product platform for Fae, an electric vehicle rental startup, during a dynamic summer internship.
• Integrated Razorpay payment gateway, facilitating seamless transactions for thousands of users.
• Implemented Google Maps API for real-time tracking of over 1,700 electric vehicles in the fleet.
• Redesigned mobile app navigation using React Native, enhancing user engagement and experience.

Jun 2021 – Aug 2021 · 2 mos · Delhi, India

Jun 2020 – Jul 2020 · 1 mo · Delhi, India