π B.Tech CCE @ Manipal University Jaipur (2022β2026)
πΌ SDE Intern (AI/ML & Backend) @ DigiFortex | Ex-Adani Green, Celebal, Adani Power
π¬ Building production-grade AI pipelines at the intersection of technology, strategy & execution
I'm a Data Scientist and AI Software Engineer who architects AI-driven product systems that transform fragmented, real-world data into structured, decision-ready intelligence for enterprise use cases. Currently at DigiFortex and previously worked at Adani Green Energy, Celebal Technologies, and Adani Power.
- π€ LLM & RAG Systems: Built AI-enabled compliance pipelines with intelligent document extraction, classification, and LLM-powered reasoning layers
- βοΈ Cloud-Scale Engineering: Hands-on with AWS, GCP, and Azure β worked with large-scale datasets and production pipelines on GCP
- π Data-Driven Impact: Time-series forecasting models with 15% accuracy improvement and ML automation reducing manual effort by 40%
- ποΈ Product Thinking: Competitor benchmarking, market gap analysis, and translating business requirements into scalable system design
- π Proven Results: Dashboards reducing debugging latency by 70% and engagement boosts of 18β20%
π Cursor-like Intelligence | Multi-Mode AI Assistant | Repository Intelligence
Advanced AI-powered code analysis platform that provides Cursor-like intelligence for understanding, analyzing, and improving codebases. Features context-aware chat, security scanning, code visualization, and real-time analysis with Groq and Gemini AI.
π οΈ Tech Stack: Python, FastAPI, React, TypeScript, RAG, Vector Search, Groq, Gemini π Impact: Intelligent code suggestions, 50MB+ repository processing, real-time AI assistance
π 20% CTR Boost | 12% Churn Reduction | 500K+ Records
Comprehensive analysis of Instagram user engagement patterns using SQL and Power BI. Created actionable insights that led to significant business improvements.
π οΈ Tech Stack: SQL, Power BI, Data Analysis π Impact: 20% increase in ad CTR, 12% reduction in user churn
π’ Used at Adani Power | 40% Manual Work Reduction
Developed a real-time people counting system using OpenCV and MobileNet SSD. Successfully deployed at Adani Power, reducing manual monitoring workload by 40%.
π οΈ Tech Stack: Python, OpenCV, MobileNet SSD, Computer Vision π Impact: 40% reduction in manual monitoring at Adani Power
π 15% Accuracy Improvement | Time Series Analysis
Built an LSTM-based time series model for stock price forecasting. Achieved 15% improvement in prediction accuracy compared to baseline models.
π οΈ Tech Stack: Python, TensorFlow, LSTM, Yahoo Finance API π Impact: 15% improvement in stock prediction accuracy
β‘ Real-time NLP | 18% Engagement Boost | Crisis Response
Live dashboard using VADER sentiment analysis for market intelligence. Enabled faster crisis response and achieved 18% boost in user engagement.
π οΈ Tech Stack: Python, Streamlit, VADER, NLP π Impact: 18% engagement boost, faster crisis response
π 70% Debugging Time Reduction | Slack Integration | 1000+ Logs
Real-time log analyzer with Slack alerts and anomaly detection. Processes 1000+ log entries with service-wise filters and 10-second auto-refresh.
π οΈ Tech Stack: Python, Streamlit, Slack API, SQLite π Impact: 70% reduction in debugging latency, <2s Slack alert response