ASX CHESS Replacement Program
- Reduced consumer lag 75% and sustained sub-200ms p99 latency by engineering high-volume Kafka event-driven services processing 5M+ events/day with exactly-once delivery via idempotency keys and manual offset commits, powering a multi-channel notification platform.
- Drove large-scale system design and end-to-end delivery of 4 distributed microservices on the ASX CHESS Replacement - a national clearing platform serving 100+ market participants benchmarked at 20M+ trade settlement events/day.
- Served as ASX client's primary technical contact over a 14-month cross-geo engagement (Chennai–Sydney); independently delivered 25 user stories, resolved 73 client-reported defects through UAT cycles, and shipped SAML SSO integration with zero-downtime operations.
- Cut average API response latency 40% (from ~800ms to ~480ms) by designing and building 15+ scalable REST APIs in Java and Spring Boot with query indexing, pagination, and performance tuning - consumed by internal platform services and external broker-facing systems.
- Eliminated recurring JVM OutOfMemory failures across 60+ production deployments in 30+ environments by building a Kafka stress-testing framework benchmarking throughput, consumer lag, and latency with Prometheus and CloudWatch.
- Mentored and led a 4-engineer team through code reviews, design reviews, and production release ownership; 2 mentees earned TCS internal awards.
- Saved ~15 hours per rollout by building 6 shell-script automation tools for SAML, SCIM, and TLS configuration; automated 600+ system-code SQL queries and drove adoption across 4 teams.
