Backend engineer with 5 years shipping production systems across IT automation, event-driven architectures, and agentic AI. I build scalable backend pipelines, multi-agent frameworks, and robust orchestration tools.
Python, JavaScript, SQL, React.js, Vue.js, FastAPI, Flask, LangGraph, RAG, LLM, Software Architecture, AWS, Docker, RabbitMQ, MCP
Architected event-driven order processing engine consolidating orders from multiple retail partners via EDI vendors into dual ERP systems and distributed WMS/3PL networks using loosely-coupled modular architecture with real-time visibility
Built AI-powered customer service assistant using LangGraph and Retrieval-Augmented Generation (RAG) with multi-agent orchestration to draft context-rich replies and automate workflows
Optimized marketing messaging platform with time zone management and regulatory compliance for global market expansion
Designed proprietary Python library with CI/CD pipeline featuring database operations, logging management, and Slack integrations for standardized internal tooling
Drove major enhancements to queue-based real-time network monitoring & IT automation platform, improving visibility, health checks, and integrations (ServiceNow, VMware, SSH, SNMP), delivering 40% faster incident response and 40% less manual effort
Designed automation exchange platform eliminating 40K+ manual hours annually and cutting client OpEx by 30%
Developed Python shell interface to run custom commands, boosting operational efficiency by 20%
Built drag-and-drop interface for orchestrating automations, reducing incident resolution time by 50%
Developed Customer Advocacy tracking tool for managing customer-developer details, feature enhancements, activities, issues, patch history, and build information using VueJS, MongoDB, and Flask, resulting in 30% decrease in miscommunication errors
Designed product health tracking tool that monitors product health across different build versions based on bugs and support tickets logged in Jira, resulting in 40% reduction in bugs in new releases
Built developer story tracker chart using VueJS for visualizing and tracking development stories across sprints
Next-generation hotel booking platform that solves the OTA visibility gap using conversational AI — users describe complex preferences in natural language and agents extract structured search parameters across multi-turn dialogue. Implements a proof-of-concept Agent-to-Agent (A2A) communication protocol where client-side and hotel-side AI agents negotiate directly for real-time availability and instant booking confirmation, eliminating intermediary commission structures. Built with production-grade patterns: stateful conversation management, dynamic generative UI that adapts components to conversation context, and live third-party hotel inventory integration with a functional demo booking flow.
AI-powered MCP server that automates git workflows by generating conventional commit messages from staged changes and creating detailed PR descriptions with customizable templates Features token-aware handling for large diffs with pagination support and ensures safety by only working with staged changes to prevent accidental commits
Personal automation project built while shopping for my own BMW to monitor inventory for exact color, package, and price combinations without daily manual checks Schedules Playwright-based token fetching, queries BMW APIs, and layers API-level plus client-side filters to pinpoint vehicles that match strict preferences Tracks seen vehicles in SQLite, exports curated Excel sheets, and only emails when brand-new preferred listings appear so nothing slips by
Comprehensive web application for managing badminton leagues and tournaments, enabling users to browse leagues, track team standings, view match statistics, and monitor rankings Interactive leaderboard with team performance metrics and detailed match history featuring win/loss visualizations using ECharts for dynamic data visualization Real-time search across leagues with league registration and team analytics featuring expandable match details for comprehensive performance tracking Built with Vue.js and Vuetify for a responsive Material Design UI, integrated with Python/Flask REST API backend and MongoDB Responsive design across desktop, tablet, and mobile devices contributing to a 50% increase in user engagement