At Easebuzz, I architected a GST invoicing and reconciliation platform that processes 1.5M+ daily transactions for 20K+ merchants. It replaced an eight-hour manual process with a five-minute pipeline, and idempotent commands made retries safe while eliminating duplicate invoices.
I also built AI-enhanced observability workflows that summarize CloudWatch logs into incident reports, along with a LangChain text-to-SQL interface for data requests. These tools cut root cause analysis time to 30 minutes.
For payment operations, I built a postpaid settlement engine and DRF payout APIs with Redis-backed idempotency keys and circuit breakers. I also led a live Blue-Green migration of payment services from Python 2.7 to Python 3.10 and Django 5.0, eliminating CVEs and boosting throughput by 42%.
Before Easebuzz, I was a Solution Analyst at Deloitte Touche Tohmatsu Limited, where I performed SQL-based data extraction and transformation across banking systems and supported SOC and SOX compliance efforts. My work has focused on reliable fintech systems, including reconciliation, settlement, and failure handling.
