Architecting AI-driven multi-agent systems and building scalable full-stack platforms with Google ADK, FastAPI, and RAG pipelines for intelligent data analysis and visualization.
Key Achievements
- Architected Hybrid Loop multi-agent system (8 agents) using Google ADK with deterministic orchestration, intent classification, and sub-agent routing, cutting avg response time from 12s to 4s
- Shipped natural-language-to-SQL pipeline with RAG-powered Vega-Lite generation using vector databases, self-critique Reflection Agent, and parallelized metadata fetching, achieving 95%+ chart accuracy with zero timeouts
- Reduced perceived wait time by 3x by delivering a Server-Sent Events streaming endpoint with typed status events, replacing static loaders with live agent progress tracking
- Designed structured prompt templates and evaluation harnesses to benchmark intent classification accuracy at 92% across multi-turn conversations
Technologies Used
Google ADKFastAPIPythonReact.jsshadcn/uiRAGLLM RoutingReflectionVector DatabasesVega-LiteSSE
Led development of enterprise-grade web applications with React, Redux, and TypeScript, implementing interactive dashboards and improving development workflows.
Key Achievements
- Led development of a licensing dashboard using React, Redux, and TypeScript with RESTful APIs, replacing manual Excel workflows and reducing overhead by 40%
- Cut initial payload by 35% by implementing interactive Plotly.js dashboards in Next.js with lazy loading, code splitting, and responsive layouts across desktop and mobile
- Designed a drag-and-drop folder management system, improving UX and reducing navigation time by 30%
- Established test-driven development workflows through bi-weekly code reviews, contributing to a 20% drop in QA bugs
Technologies Used
ReactReduxTypeScriptNext.jsPlotly.jsREST APIsJestRTL
Delivered production-ready frontend features with modular components, comprehensive test coverage, and performance optimization.
Key Achievements
- Built a low-code content management system using Strapi, React, and GraphQL with role-based access control, cutting content turnaround time by 50%
- Automated Okta app creation and auth middleware setup via Node.js scripts, reducing app provisioning time by 70%
- Increased test coverage from 20% to 80% using Jest and React Testing Library, leading to a 25% drop in post-release bugs
- Improved Lighthouse score from 75 to 92 by migrating to modular TypeScript components with code splitting, tree shaking, and lazy route loading
Technologies Used
ReactTypeScriptGraphQLStrapiNode.jsOktaJestRTL
Completed Bachelor's degree in Information Technology with a focus on software development and engineering fundamentals.
Key Achievements
- Graduated with GPA of 9/10
- Developed strong foundation in algorithms, data structures, and software engineering principles
- Completed multiple academic projects in web development and application design
- Participated in technical workshops and coding competitions
- Built foundation in object-oriented programming and software architecture
Technologies Used
JavaC++Data StructuresAlgorithmsObject-Oriented DesignWeb DevelopmentDatabase Systems