Technology Lead at Amazon · Founder of Assevra AI · IEEE-published researcher
Agentic AI & Distributed Systems Engineer
I build reliable, large-scale agentic AI and distributed systems — and research how to make autonomous software trustworthy in production. Eleven-plus years spanning enterprise B2B commerce, cloud platforms, and low-latency trading.
What I work on
Professional focus
Agentic & Autonomous AI
LLM orchestration, multi-agent architectures, and developer copilots that plan, act, and self-improve within governed boundaries.
Trustworthy & Secure GenAI
Reliability scoring, guardrails, evaluation harnesses, and hallucination-risk validation for autonomous systems in production.
Distributed & Event-Driven Systems
High-throughput, low-latency, fault-tolerant platforms built on resilient messaging and cloud-native infrastructure.
B2B Commerce & eProcurement
Enterprise ordering, ERP and marketplace integration, multi-entity procurement, and resilient supply-chain execution.
AI-Assisted Software Engineering
Autonomous CI/CD, release-safety automation, and AI-augmented code review that raise developer velocity without lowering the bar.
Cloud-Native Platforms
API design, service reliability, and operational and cost optimization across large-scale AWS environments.
Selected research
Publications
CostAgent: Self-Improving Autonomous LLM-Based Orchestration for Cost-Optimal Cloud Data Processing at Scale
A self-improving, LLM-based orchestration approach for cost-optimal cloud data processing at scale.
Multi-Agent Debate for Software Architecture Governance: A Framework for ADR Generation, Risk Review, and Deployment Readiness
A multi-agent debate framework for software architecture governance — automating ADR generation, risk review, and deployment-readiness assessment.
Open source
Selected project
Assevra AI
An MIT-licensed, open-source reliability scorecard for LLM agents. Assevra AI scores agent reliability against fixed thresholds so teams can measure — not guess — whether an agent is safe to ship.
Beyond the papers
Writing & speaking
Claude Code Doesn't Ship Production Code: Your Config Does
Why the configuration around an AI coding agent — not the model — determines whether it ships production-ready code.
Keynote · 2026 IEEE 5th World Conference on Applied Intelligence and Computing (AIC 2026), Jabalpur, IndiaFrom Autonomy to Assurance: Governing Agentic AI for Production Enterprise Systems
Keynote speaker. IEEE Xplore / SCOPUS-indexed proceedings.
Let's talk
Open to research collaboration, speaking, technical advising, and conversations on agentic AI reliability.
Get in touch