Intelligent Data Deduplication & Clustering Platform
ML system that detects and clusters duplicate records across a 50K+ document corpus, with a feedback loop that retrains as reviewers correct it — cutting duplicate records by 80%.
Building scalable Python backends, AI/LLM systems, and cloud-native infrastructure.
Senior full stack & AI engineer with 4+ years shipping production software on AWS & Azure — from microservices and ETL pipelines to LLM-powered applications. Based in Ahmedabad, India.
01 — selected work
Ten production systems spanning AI/LLM applications, data engineering, and full-stack web platforms.
ML system that detects and clusters duplicate records across a 50K+ document corpus, with a feedback loop that retrains as reviewers correct it — cutting duplicate records by 80%.
Production-grade data platform on Microsoft Fabric using Medallion Architecture and PySpark transformations, delivering analytics-ready data for enterprise Power BI reporting.
Collaborative authoring tool that modernises ageing book content with generative AI — styling, enhancement, and publishing in one workflow. Cut content production time by 65%.
Real-time activity tracker consolidating metadata from dozens of tools, with semantic search and AI duplicate detection surfaced through a single dashboard.
Real-time football data and predictive insights for league agencies, powered by ETL pipelines and match analytics.
Multi-tenant education platform built on microservices for authentication, administration, content delivery, and safety compliance.
SaaS workforce system covering leave tracking, project management, and employee activity, running on AWS.
Ride-booking platform for the Spanish market with live driver tracking and admin revenue reporting.
Predictive model that flags high-risk stroke patients early, using SMOTE oversampling to handle imbalanced clinical data.
Scraping pipeline over the Play Store catalogue with cleaning and normalisation, producing a reliable app-ecosystem insights database.
02 — background
Leading teams from architecture to production deployment, across cloud, AI, and web platforms.
03 — toolkit
60+ technologies across languages, frameworks, cloud, AI, and data — grouped by where they belong.
04 — education
05 — contact
Open to full-time roles, freelance projects, and consulting — especially AI-augmented products and data platforms.