I am Pranit, a Full-Stack AI Developer with a Civil Engineering background. I build AI-powered products that solve real-world domain problems — from Vastu-compliant floor plan generators to construction cost estimation tools. I bridge the gap between physical world domain knowledge and modern AI systems.
A timeline of impact-driven roles where I've leveraged data and AI to solve complex business problems.
AI-powered floor plan generator that creates Vastu-compliant architectural drawings from a text prompt. Features BSP layout engine, room-by-room compliance scoring, 3D visualization, fire evacuation pathfinding (A*), crowd simulation with CO₂ analysis, sun path analysis, seismic risk mapping (IS 1893:2016), and AutoCAD DXF export.
Live AI-powered construction cost estimator platform backed by pan-India government DSR/SSR rates. Features an autonomous Estimator agent that generates structured BOQs from natural language project descriptions, a Rate Finder for instant semantic search across state SSR and central CPWD DSR data, automated geo-specific escalation, temporal auditing against gazette versions, Excel export with measurement book, and an Enterprise tier for org-wide multi-project portfolio management. Website live and in active testing.
Engineered a machine learning pipeline to predict loan defaults, improving risk assessment accuracy. Deployed as an interactive web app using Streamlit.
Architected an end-to-end data pipeline to track volatile material costs for civil engineering projects using SQL and Power BI.
Queried 1M+ rows of housing data in BigQuery to identify undervalued investment opportunities and visualized regional price trends.
I believe that great engineering comes from a diversity of inputs. Here's what keeps my mind sharp.