"Machines don't just need data — they need direction."

I'm a Computer Science student and ML Engineering Intern passionate about building applied Machine Learning, MLOps, and Generative AI systems that solve real world problems — from industrial predictive maintenance to full-stack SaaS products.

I enjoy designing end-to-end ML pipelines, from feature engineering and model explainability to deployment and drift monitoring, while also building the full-stack applications that put them in front of real users. I'm constantly exploring the latest in LLMs, AI agents, and cloud infrastructure, and I stay active in Bengaluru's tech community through hackathons, meetups, and mentorship programs.

I believe in continuous learning, proof-of-work over credentials, and building technology that creates measurable impact.

• 🏢 EstateFlow — Property management SaaS with a custom RBAC engine and account-level data isolation
• 🤖 Parallel multi-agent AI systems with shared memory layers
• 📚 12-week Gen AI Engineering path — RAG, agents, LoRA/QLoRA fine-tuning
• ☁️ Cloud-based ML deployment on AWS

Also Experienced With

• Scikit-learn
• XGBoost
• SHAP (Explainability)
• Pandas / NumPy
• MLflow / DVC
• Prompt Engineering
• AI Agents

Industrial predictive maintenance pipeline for real-time equipment failure prediction using sensor telemetry.

Key Features

• ~110 engineered features from raw sensor data
• XGBoost-based failure prediction
• SHAP explainability for model transparency
• CUSUM drift detection for data quality monitoring
• Real-time ingestion via MQTT/HTTP

Tech Stack

Python • XGBoost • SHAP • Flask • MQTT

AI-powered water quality monitoring platform built at the GDG "Build for Bengaluru" hackathon, combining IoT telemetry with ML-based forecasting.

Key Features

• Real-time IoT sensor telemetry
• ML-based water quality forecasting
• Interactive monitoring dashboard
• REST API integration layer

Tech Stack

Next.js • FastAPI • Flask • IoT

Property management SaaS designed for organizations to manage properties, users, and operations through a scalable, account-isolated architecture.

Key Features

• Custom Role-Based Access Control (RBAC) engine
• Account-level data isolation
• Glassmorphism dashboard design
• Authentication & Authorization
• Task & Property Management

Tech Stack

Next.js (App Router) • React 19 • TypeScript • MongoDB

LSTM-based Remaining Useful Life prediction model for predictive maintenance applications.

Key Features

• LSTM sequence modeling on multivariate sensor data
• RMSE ~12 on FD001 dataset
• Asymmetric loss function experimentation

Tech Stack

Python • PyTorch • LSTM

• Machine Learning & MLOps
• Generative AI & AI Agents
• Applied ML Systems
• Full-Stack Development
• Explainable AI
• Cloud & Deployment
• Open Source

If you like my work, consider starring my repositories ⭐

Thanks for visiting my profile — feel free to explore my repos and connect with me.