Mandates - Automatic Payments Platform
Redesigned a high-throughput financial mandate system to a robust, event-driven architecture, significantly improving scalability and reliability.
Scroll to zoom • Drag to pan • Click fullscreen for more options
Impact
- 3x improvement in execution latency and throughput
- 10+ tenants onboarded with isolated and reliable processing
- Zero incidents of double execution or concurrency-related failures
- Scalable architecture aligned with stringent financial transaction deadlines (e.g., 2 PM cutoffs for Mutual Funds)
Key Contributions
- Architected a new mandate schema with denormalized fields to enable high-performance querying and reduce database overhead.
- Replaced Cron-based execution with Dynaq, a distributed event scheduler, enabling asynchronous and resilient task orchestration.
- Built an Event Tracker to trace mandate, transaction, and execution lifecycles, dramatically improving observability and debuggability.
- Enforced multi-tenancy and tenant isolation, ensuring fault containment and SLA guarantees across all clients.
- Implemented global locks and burst control to handle peak loads and race conditions during high-traffic windows.
- Introduced audit-compliant history tables to track state transitions and support traceability across workflows.
Implementation Details
Role: Backend Engineer | Lead Contributor Tech Focus: Distributed Systems, Event-Driven Architecture, Multi-Tenancy, Performance Optimization Redesigned a high-throughput financial mandate processing system to address performance bottlenecks, scalability limits, and operational fragility in the legacy platform. Transitioned from a Cron-based execution model to a robust, event-driven architecture, ensuring reliable, low-latency execution across 10+ tenants.
Technologies & Tools
Java
Spring Boot
Apache Kafka
Dynaq
Distributed Systems
Event-Driven Architecture
Multi-Tenancy
Performance Optimization