Staff Software Engineer — Payments Risk, Fraud Platforms & Agentic AI
Portfolio: santanusetu.github.io
Staff Software Engineer at Visa Inc. with 10+ years building global payments, risk and fraud platforms. I lead platform, observability and agentic AI initiatives on Visa Risk Manager — the fraud-decisioning platform 1,000+ issuing banks use to approve or decline transactions in real time.
Before Visa: Infosys Ltd. (Senior Systems Engineer, 2011–2014) delivering Java and web features for EA Sports' Origin platform and Android set-top-box capability for Comcast's Skype on XFINITY, and a software engineering internship at Walmart Global Tech (@WalmartLabs) building one of Walmart's first driver-facing delivery apps.
M.S. Software Engineering, San Jose State University · B.E. Computer Science, Visvesvaraya Technological University.
| Project | What it does |
|---|---|
| Enterprise MCP Server | Model Context Protocol server exposing rule, KPI and reporting data as permission-scoped tools, letting AI agents act on governed production fraud data inside enterprise guardrails. Adopted by 6 teams. |
| Agentic AI Security Remediation | Remediation framework built on compounding-engineering principles. Closed 130 of 130 security findings and cut per-finding remediation from a day to 45 minutes; produced 19 patterns and 5 workflows now standard across 40+ repositories. |
| SLA Observability Platform | Observability across 8.34M+ monthly requests on production endpoints. Cut SLA-miss detection from three weeks to under a day. |
| React Micro-Frontend Migration | Migrated 85+ Angular pages to React and TypeScript micro-frontends on Module Federation with zero production incidents, improving feature lead time by 35%. |
- Guest Lecturer, Stanford University (2×, 2025–2026) — representing Visa in a Visa–Stanford collaboration on payments fraud and risk systems.
- Best Financial Crime Prevention Solution, Pay360 Awards 2023 — awarded to the Visa Risk Manager team.
- TD Commons Publication, 2021 — co-authored "System and Method for Monitoring Fraudulent Transactions at Merchant Level in Real Time Payment." Read it
Backend & distributed systems — Java, Spring Boot, Microservices, REST APIs, Kafka, Redis, Python, Go AI, MCP & agentic engineering — Agentic AI, Model Context Protocol (MCP), RAG, LLM applications, tool integration Frontend & full-stack — React, TypeScript, Redux, Micro-Frontends, Module Federation, Angular Data & cloud — SQL, DB2, StarRocks, MongoDB, Docker, Kubernetes, AWS, Jenkins, CI/CD Observability — Prometheus, Splunk, SLA/OLA monitoring, HashiCorp Vault Security & testing — SAST, DevSecOps, OWASP Top 10, JUnit, Selenium, Cypress, JMeter
Open-source and academic work from 2015–2016, kept for reference: iShare · MovieBuddy · Mobile test automation · Amazon EC2 monitoring · VM disaster recovery
The source for santanusetu.github.io — a static
single-page site served by GitHub Pages. No build step: index.html, css/, js/
and img/ are deployed as-is.
- Email — santanu.setu@gmail.com
- LinkedIn — santanu16
- GitHub — santanusetu
- Stack Overflow — santanu
- Résumé — download PDF
I am currently exploring new opportunities in platform engineering and agentic AI.