Software engineer who has designed and shipped 6 production AI agents and integrations in the past year — multi-agent RAG pipelines, autonomous QA agents, and conversational AI UX — on top of 6+ years shipping full-stack applications at Infosys/EdgeVerve.
AI Agent Developer & Full-Stack Engineer
I'm a Software Engineer with 6+ years of full-stack experience at EdgeVerve and Infosys. Over the past year I've designed and shipped 6 production AI agents and integrations — multi-agent RAG pipelines, autonomous QA and governance agents, and conversational AI UX — used by real engineering teams, not tutorials.
My AI work focuses on embedding intelligence into existing systems — not building demos. I design multi-agent RAG pipelines, LLM-powered middleware, and context-bounded assistants that operate inside real production constraints. AI as an integration layer, not a replacement.
6 production AI agents & integrations shipped in the past year — multi-agent RAG pipelines (AI Canvas, Claude/Bedrock, PGVector), autonomous QA/governance agents, and conversational AI UX
Building end-to-end enterprise applications with React, Angular, and modern JavaScript frameworks
Designing scalable APIs and microservices with Spring Boot, Node.js, and Express
Bengaluru, India
Bengaluru, India
India
India
Organized by category. Each skill is tied to real production work — not self-assessed percentages.
Primary focus — applied in production and internal agent tooling
Shipped in enterprise products, used by real clients
Working knowledge, used in production or agent tooling
Built a 4-agent RAG pipeline (retriever → analyst → creative director → storyboard engineer, plus a QA validator) that turns enterprise documentation into narrated, animated explainer videos — with a from-scratch client-side animation and MP4 export engine.
Built a zero-dependency Python agent that matches error signatures against curated playbooks and cross-references live GitHub code search across linked repos — turning a manual, multi-repo debugging routine into a single query.
Authored an autonomous agent that crawls documentation portals with Playwright, validates every checklist item with an LLM, and generates a shareable HTML/JSON audit report with confidence-scored verdicts and evidence screenshots.
Designed a custom VS Code agent that classifies every request and routes it to the cheapest safe model tier based on remaining Copilot budget — keeping full agentic capability while enforcing cost discipline across the team.
Added a Google-AI-Overview-style GenAI answer experience to an enterprise MkDocs portal, and built the context-engineering layer behind a conversational banking UI — summarizing full user journeys into compact context sent with every AI query.
The site you're on right now — including 3 working AI agents (chat, resume fit analysis, code review) built with the same constraint-first, structured-output patterns used in production.
Built full-stack UI and APIs across three Finacle banking modules — transaction tracking, expense approval workflows, and cross-border payment processing — serving 50+ enterprise banking clients.
Three working agents built on this site — chat, resume fit analysis, and code review — all sharing the same constraint-first, structured-output patterns I use in production. View source on GitHub
This chat uses Gemini's function-calling API with 4 custom tool definitions. The LLM decides which tools to invoke based on conversation context. My entire profile is injected as structured context — the same pattern I use for production SDK documentation assistants.
This tool sends the job description along with my full profile to Gemini and requests a structured JSON response covering: skill matching, fit scoring, gap identification, and a tailored summary — the same structured-output pattern I use in production AI integrations.
This agent demonstrates constraint-bounded AI and structured output enforcement. It's constrained to only review code, scores across 5 quality dimensions, and returns a strict JSON schema that the UI renders into visual results.
Sri Krishnadevaraya University
Feel free to reach out for collaborations, opportunities, or just a friendly conversation about technology and AI.