I'm a Software Development Engineer at Visa Inc. in Bengaluru, contributing to Open VisaNet — Visa's next-generation payment processing platform. My work involves designing and building authorization features in Go that achieve functional parity with decades-old mainframe systems. You know that split second between tapping your card and the relieved beep of approval? Yeah — I build that. (And occasionally, the declined one too.)
I graduated from College of Engineering, Guindy (Anna University) in 2024 with a degree in Information Technology and a CGPA of 9.22/10. Outside of production systems, I spend time on applied AI — RAG pipelines, fine-tuning, and tools that reduce the gap between raw LLM capability and real engineering workflows.
I also write occasionally about the things I'm learning — check out the blog.
Software Development Engineer
Visa Inc. · Bengaluru- Develop core authorization features for Open VisaNet (OVN), Visa's next-generation payment processing platform — including velocity fraud prevention and financial adjustment modules in Go, achieving 100% functional parity with legacy mainframe systems.
- Led development of a natural language-driven test case automation tool using CrewAI, Python, and Visa internal APIs, reducing manual QA effort by 50% and enabling automatic identification of test scenarios across authorization workflows.
- Awarded Top 3 in GenAI Hackathon 2025 (out of 20+ teams) for a Knowledge Retrieval Tool using RAG to consolidate fragmented mainframe documentation across Jira and Bitbucket.
- Designed an automated reporting system in Go tracking feature completion and release-readiness metrics, adopted by engineering leadership — reducing reporting time by 80%.
Software Development Intern
Visa Inc. · Bengaluru- Developed a POC to parse, clean, and transform Visa's proprietary transaction profile data using a Python CLI. Ingested 10,000+ profile entities into Elasticsearch via Logstash, reducing search time from 5+ minutes to under 10 seconds.
OEM Data Simulator
↗
Realistic vehicle telemetry simulator for the Motorq Hackathon —
emulates speed, fuel, GPS, and tire pressure across diverse
driving scenarios using the SUMO engine and TraCI API.
Plug in your traffic conditions, routes, and topography, hit
go, and out comes realistic vehicle telemetry at whatever
sampling frequency you need — no actual car required.
AI Architectural Workflows
↗End-to-end AI system generating 2D floorplans and photorealistic interior designs from natural language prompts. Evaluated GPT-3.5, Llama 3, and Mistral. Published in IEEE CVMI 2024.
Real-Time Quiz App
↗
Microservice-based quiz platform with a Go backend, goroutines,
JWT authentication, PostgreSQL storage, and WebSocket support
for real-time updates. React/TypeScript frontend.
At work, we toss around terms like Microservices, Kubernetes,
gRPC, and Docker like they cost nothing — because a brilliant
DevSecOps team quietly abstracts all this away. This project
was a way for me to understand all this on a deeper level:
turns out, making distributed systems not fall apart at 2 AM
is genuinely hard work :)
Languages
AI & Machine Learning
Frameworks & Libraries
Infrastructure & Tools
Bachelor of Engineering — Information Technology
College of Engineering, Guindy · Anna University