Hi, I'm Kiran U Kamath

Lead Software Engineer specializing in high-performance backend systems, microservices architecture, and fraud risk management platforms. I turn complex problems into scalable, low-latency solutions.

Get in touch View my work
Scroll
01About

Get to know me

Kiran U Kamath KU

I'm a Lead Software Engineer based in Bengaluru with 6+ years of experience in the fintech space. I love building things that work at scale — systems that are fast, reliable, and thoughtfully designed.

When I'm not deep in code at PayU, I'm usually working on a side project, exploring agentic AI and distributed systems, or finding better ways to solve old problems. I believe great software is as much about clarity of thought as it is about technical skill.

Role
Lead Software Engineer
Company
PayU
Based in
Bengaluru, India
Focus
Fraud Risk Platforms
LinkedIn GitHub Resume
02Career

Work Experience

Apr 2026 – Present
Bengaluru, IN

Lead Software Engineer

PayU
  • Team Leadership: Lead a team of 7 engineers across the Fraud Risk Management platform, owning the AOP roadmap and driving technical prioritization, quarterly planning, and delivery.
  • Agentic AI & Production Stability: Designing agentic AI use cases and workflows, improving observability, reducing incidents, and establishing engineering standards across the team.
Apr 2023 – Mar 2026
Bengaluru, IN

Senior Software Engineer

PayU
  • System Re-Architecture: Spearheaded migration of the Trident FRM system from a monolithic Java app to a scalable, multi-tenant microservices architecture using Spring Boot, RabbitMQ, and Couchbase — delivering significant cost reductions and performance improvements.
  • Rule Engine: Designed a rule engine capable of processing 5× the transaction volume of the previous version, supporting dynamic rule authoring and real-time execution without redeployment.
  • Real-Time Statistics: Developed an optimized real-time statistics computation model for rule evaluation using Couchbase as an in-memory database for low-latency data fetch and storage.
  • Workflow Automation: Designed workflows for fraud detection and AML using Camunda BPMN for multi-level approvals and process orchestration.
  • Performance Optimization: Reduced decision engine latency from 630ms → 50ms via Java Profiler and memory leak analysis. Scaled throughput from 40 TPS → 200 TPS per node (4-core, 8GB RAM VM).
  • Feature Development: Implemented case management workflows, real-time currency conversion, suspicious entity detection, and entity profile enrichment.
  • Engineering Excellence: Led code reviews enforcing SOLID principles, design patterns, and an API-first approach across the team.
Apr 2021 – Apr 2023
Bengaluru, IN

Software Engineer

PayU
  • Reporting Service: Delivered a fully customized reporting service powered by Couchbase Analytics, reducing TAT from hours to minutes while offloading operational database load and supporting ad-hoc queries.
  • Rule Engine – Case Manager: Contributed to FRM Trident 2.0, enhancing rule engine capabilities and building the case manager portal for fraud monitoring.
  • DBInsert Microservice: Designed a robust DBInsert microservice using RabbitMQ, Kafka, MySQL, and Couchbase for seamless data transformation and integration between messaging systems and databases.
  • Data Migration Utility: Built a migration utility to transfer data from Couchbase to MySQL using Kafka and Maxwell's Change Data Capture, improving data consistency and scalability.
Sep 2020 – Apr 2021
Bengaluru, IN

Associate Software Engineer

PayU
  • Worked on ML model development for RBA Scorecard and Fraud ML models — including hypothesis testing, feature engineering, and hyperparameter tuning.
  • Integrated ML models into the FRM Rule Engine.
  • Contributed to feature development, bug fixes, and production issue resolution.
03Toolbox

Skills

Languages

Java Python SQL Shell

Frameworks & Tools

Spring Boot Hibernate Camunda BPMN REST APIs Java Profilers

Databases & Caching

Couchbase (NoSQL) MySQL Redis (Lua Scripting) Elasticsearch

Messaging & Streaming

RabbitMQ Apache Kafka

AI & Agentic Systems

Agentic AI Workflows LangGraph LangChain AWS Bedrock RAG Prompt Engineering

DevOps & Cloud

Docker Kubernetes AWS (EC2) CI/CD

Practices

API-First Design SOLID Principles Design Patterns Code Reviews
04Side projects

Things I've built

KiraDB

A Redis-compatible database with adaptive tiered storage (memory → SSD → S3). Features built-in feature flags, rate limiting, config management, and semantic caching designed for AI workloads — all in one system.

Distributed DBRedis-compatibleAI InfraTiered Storage
View on GitHub →

Ticket Booking Backend

Designed and developed a high-concurrency ticket booking system handling parallel bookings efficiently. Uses Redis as a distributed cache with Lua scripts to manage seat availability and prevent overselling.

JavaRedisLua ScriptsConcurrency
View on GitHub →

Loop

Agentic interview coach built on LangGraph. Covers end-to-end interview prep — planning, mock interviews, grading, and adaptive learning across sessions using multi-agent workflows.

Agentic AILangGraphAWS BedrockPython
View on GitHub →
05Writing

Recent Writing

Apr 2, 2026
Redis Hashes: Memory Optimization at Scale
How Redis Hashes reduce memory overhead by consolidating keys and leveraging ziplist compression — with real numbers and a gaming leaderboard use case.
RedisBackendPerformanceSystem Design
View all posts & LinkedIn writing →
06Background

Education

2016 – 2020

B.E. – Information Science & Engineering

National Institute of Engineering

GPA: 8.24 / 10

2023

Cloud Native Application Architecture

Udacity

Certificate

2021

Data Scientist Nanodegree

Udacity

Nanodegree

07Get in touch

Contact

Feel free to reach out — I'm always open to interesting conversations and new opportunities. Email or LinkedIn is the fastest way to reach me.

Email
kirankamat.mgm@gmail.com
LinkedIn
kiranukamath
GitHub
kiranukamath
Resume
View / Download