AI Engineer building applied AI, ML, and analytics products for construction and field operations.
Assistant Professor at the University of New Haven
Ph.D. in Civil Engineering | M.S. in Data Science (Expected Spring 2026)
I build AI systems for construction safety, inspection, document intelligence, and operational decision support. From Civil Engineering to Construction Management, an MBA, and now Data Science, my journey reflects how much I love learning, adapting, and growing across disciplines.
- Applied AI and ML product development
- Multimodal AI for image, document, and field workflows
- Predictive modeling and explainable risk analytics
- Document intelligence and grounded LLM applications
- Construction safety, inspection, and operational decision support
AI-powered inspection reporting app that turns field photos into structured defect findings, GPS-aware map context, and report-ready summaries.
What it shows: multimodal prompting, structured outputs, EXIF GPS extraction, map-aware review, and AI-assisted reporting workflows.
Tools: Python, Streamlit, OpenAI Responses API, Pydantic, Pillow, pandas, geopy
AI document Q&A copilot for construction specs, safety manuals, and method statements with grounded answers, summaries, and source excerpts.
What it shows: grounded LLM workflows, document parsing, structured answer design, source-backed responses, and report generation.
Tools: Python, Streamlit, OpenAI Responses API, Pydantic, PyPDF, python-docx
Interactive ML dashboard for predicting schedule delays and budget overruns from project planning data.
What it shows: supervised learning, risk scoring, what-if analysis, explainability, and product-style ML presentation.
Tools: Python, Streamlit, scikit-learn, pandas, NumPy, Plotly
Interactive SQL dashboard for analyzing construction near misses, incidents, hazards, and corrective actions.
What it shows: relational data modeling, SQL analytics, KPI design, trend analysis, and operational dashboarding.
Tools: Python, SQLite, SQL, pandas, Streamlit
Multimodal safety copilot for analyzing site images and generating structured observations, follow-up questions, and report-ready outputs.
What it shows: vision-language workflows, safety-oriented AI assistance, structured reporting, and applied product thinking for field teams.
Tools: Python, Streamlit, OpenAI API, multimodal prompting, structured outputs
- More projects: GitHub Repositories
- Publications and research: Google Scholar
I have worked on many more AI and engineering projects than what is visible here. Some of that work cannot be shared publicly because of funding restrictions, ownership, collaboration agreements, or client constraints. I still keep GitHub updated with representative public projects that reflect the kinds of systems I build and the problems I enjoy solving.
I make one of the best baklavas in the world, and yes, precision matters everywhere.