Designing AI systems
that ship to production.

Cincinnati, OH

• Engineered an autonomous Service Parking module with LangGraph's stateful agent graphs, persistent PostgreSQL checkpointers, and human-in-the-loop gates — cut idle compute costs up to 30%.
• Transitioned the parking system into a proactive, guardrailed Autopilot framework supporting long-running conversations that survive restarts.
• Architected an AWS Recommendation Agent on Amazon Bedrock + MCP; built a custom MCP client to AWS Cost Explorer and Jira MCP servers for closed-loop reasoning — 85% less manual analysis time.
• Built Ask-FI, a multi-agent platform using LangGraph supervisor–worker pattern with Snowflake Cortex + custom RAG (semantic chunking) — shrunk LLM context by 75%.
• Deployed Ask-FI on Docker + AWS Elastic Beanstalk with auto-scaling for enterprise availability.

Miami, FL

• Integrated ML models into a web app to deliver scalable financial insights, designed for 10,000+ users post-launch.
• Engineered a React + Next.js + PostgreSQL platform that cut development time by 50%.
• Collaborated with the founding team on an investor-ready pitch deck — helped secure $500k in funding.

Cincinnati, OH

• Analyzed AWS + Azure cloud utilization to identify optimization opportunities — reduced costs by $1.5M annually.
• Designed scalable RAG pipelines and LLM systems with LangChain — improved retrieval performance by 25%.

Pune, India

• Built RESTful microservices in Java 17 + Spring Boot with API security for a vehicle loan platform serving 170+ enterprise clients.
• CI/CD on GitHub Actions with automated quality gates — 25% team productivity gain, 10% faster releases.
• Cut API response time 35% via AJAX-based notification system, boosting engagement 20% globally.

Pune, India

• Launched Vrinda app with Spring + Maven backend and NLP-based content recommendations — boosted engagement 50%.
• Led cross-platform Flutter development consuming REST APIs for 1,000+ Android & iOS users.