Mumbai Metropolitan Region
The AI landscape is shifting from "Chatbots" to "Autonomous Agents." I build the latter. I am an AI Developer specializing in bridging the gap between raw data and automated business intelligence. Whether itโs designing multi-agent marketing pipelines or building OCR systems that cut medical processing time by 96%, I focus on high-impact AI integration. What I bring to the table: ๐๐ ๐๐ง๐ญ๐ข๐ ๐๐จ๐ซ๐ค๐๐ฅ๐จ๐ฐ๐ฌ: Designing role-based AI pipelines using ๐๐ซ๐๐ฐ๐๐, ๐๐ฒ๐๐๐ง๐ญ๐ข๐, and ๐๐๐๐ for zero-manual intervention. ๐๐ง๐-๐ญ๐จ-๐๐ง๐ ๐๐ฎ๐ญ๐จ๐ฆ๐๐ญ๐ข๐จ๐ง: Connecting Gmail, Supabase, and LLMs via ๐ง8๐ง and ๐๐๐ค๐.๐๐จ๐ฆ to eliminate manual data ingestion. ๐๐๐๐ฅ๐๐๐ฅ๐ ๐๐ง๐ ๐ข๐ง๐๐๐ซ๐ข๐ง๐ : Developing high-performance backends using ๐ ๐๐ฌ๐ญ๐๐๐ and optimizing massive datasets (32M+ records) using ๐ ๐ข๐ซ๐๐๐ฎ๐๐ค๐ฌ. Key Impact Metrics: ๐๐๐๐ฅ๐ญ๐ก๐๐๐ซ๐: Improved document processing from 50 mins to 2 mins per file. ๐๐๐๐ฅ ๐๐ฌ๐ญ๐๐ญ๐:: Achieved 94% R2 accuracy for property price prediction models. ๐๐๐ญ๐๐ข๐ฅ: Optimized data processing speeds by 50% on 32M records.
๐ง8๐ง ๐๐จ๐ซ๐ค๐๐ฅ๐จ๐ฐ๐ฌ: Orchestrated autonomous pipelines to eliminate complex manual bottlenecks. ๐๐จ๐๐๐ฅ ๐๐๐๐ฌ:: Deployed ๐๐ฅ๐ฅ๐๐ฆ๐ for high-performance, privacy-first local inference. ๐๐ซ๐จ๐ฐ๐ฌ๐๐ซ ๐๐ฎ๐ญ๐จ๐ฆ๐๐ญ๐ข๐จ๐ง: Leveraged ๐๐ฉ๐๐ง๐๐ฅ๐๐ฐ to build agents that autonomously navigate and automate web tasks.
Developed an end-to-end system called Intelligent Property Analyser that predicts housing prices using machine learning models like Random Forest and Linear Regression. I achieved a model accuracy of 90.58% with Random Forest. Additionally, I designed a Flask-based web application to allow users to interact with the model, input property details, and obtain price predictions.