I am a full stack AI engineer with experience building and deploying AI-powered products, data pipelines, and web applications from early product discovery through production delivery.
I have developed agentic document-processing systems using GPT-4o, RAG, and ReAct prompting, achieving high accuracy for contract, SEC filing, and operational-document analysis. I also have experience evaluating and improving LLM reliability with LangSmith and LangGraph.
I build backend services and financial data workflows using Python, SQL, FastAPI, Django, REST APIs, PostgreSQL, and cloud infrastructure. My work has included transforming more than 10,000 e-commerce transactions into automated accounting records and processing 15M+ monthly SKU records through ETL pipelines.
I am comfortable across the modern web stack, including TypeScript, React, Next.js, Node.js, and databases such as PostgreSQL, MongoDB, Supabase, Firebase, and Redis. I use Docker, Git, testing practices, and cloud platforms to deliver maintainable software.
I also have research experience applying retrieval-augmented generation and language models to legal, environmental, and scientific information extraction. I enjoy translating complex user and business needs into reliable, practical AI systems.
Led end-to-end MVP development, including user interviews, requirements definition, and deployment of agentic document-processing workflows.
Engineered an LLM agent with GPT-4o, ReAct prompting, and RAG that achieved 98% mismatch-detection accuracy across contracts, SEC filings, and operational files. Validated the platform with three pilot users, resulting in repeat testing and requests for a public API.
Built REST APIs and data pipelines to automate profit-and-loss and balance-sheet generation from more than 10,000 e-commerce transactions.
Translated accounting rules for transaction processing, fees, and taxes into automated categorization and reporting logic. Increased automated test coverage from 0% to 45% through unit and integration tests for financial reporting workflows.
Delivered an agent reasoning-log feature for SME financial analysis software, summarizing agent reasoning and tool interactions to improve user trust and product adoption.
Built a LangSmith-based model evaluation pipeline that enabled four engineers to diagnose and address model hallucinations. Evaluated LangGraph for agent reliability, leading to its adoption as a core roadmap component.
Conducted a pilot study on integrating RAG into an information-retrieval system for extracting water-regulation information from legal documents and environmental reports.
Developed a chatbot using BLOOM, LlamaIndex, and Streamlit that achieved 92% accuracy on water-regulation queries. Received stakeholder recognition for precision, speed, and effectiveness in regulatory compliance applications.
Deployed a Python, SQL, and Airflow ETL pipeline on GCP to process more than 15 million monthly SKU records across Kohl's departments.
Implemented unit tests and data-quality checks to validate pipeline accuracy and integrity.
Bachelor of Science, Electrical Engineering and Computer Science
AANAPISI Scholar, Osher Scholar, and Dean's List recipient. Relevant coursework included Artificial Intelligence, Computer Security, Database Management, Software Development, and Principles of Data Science.
This professional hasn’t added portfolio projects yet.
This professional hasn’t listed any services yet.
