Machine Learning Engineer | Senior Data Scientist | Generative AI & MLOps | Marketing Analytics & Product Optimization
I'm a Machine Learning Engineer and Senior Data Scientist with expertise in building scalable AI systems, deploying LLM-powered solutions, and optimizing business operations through data-driven strategies.
My work spans the full ML lifecycle — from architecting data pipelines and training models to production deployment and monitoring. I've delivered enterprise-grade solutions across healthcare, pharmaceuticals, and marketing analytics, consistently achieving measurable impact: 72% reduction in execution time, 63% decrease in memory consumption, and successful integration of Generative AI systems using Azure OpenAI.
Beyond technical execution, I lead teams, train professionals, and translate complex models into clear business outcomes. I've trained over 70 employees and managed multiple concurrent projects while maintaining high standards of code quality and system performance.
Axtria - Ingenious Insights
Axtria - Ingenious Insights
Axtria - Ingenious Insights
Axtria - Ingenious Insights
End-to-end MLOps project for predictive maintenance using engine sensor data. Includes data versioning on Hugging Face, MLflow experiment tracking, CI/CD with GitHub Actions, and Dockerized Streamlit deployment for real-time engine failure classification.
Marketing Mix Modelling app: attribute sales/revenue to channels with adstock, saturation transforms, and ROI/mROI. Streamlit wizard, 5 model types (Linear, Ridge, Lasso, Bayesian), segment analysis.
RAG-based medical Q&A over the Merck Manual (19th ed.). ChromaDB, GTE-large embeddings, Mistral 7B (GGUF). Local, privacy-first, GPU-accelerated medical assistant pipeline.
End-to-end MLOps pipeline for predicting customer purchase of a wellness tourism package. Includes data registration on Hugging Face, XGBoost model training with MLflow tracking, CI/CD using GitHub Actions, Dockerized Streamlit app, and deployment to Hugging Face Spaces.
Comprehensive learning platform for mastering Data Science, AI & ML through 250+ curated problems. Features progress tracking, dark mode, difficulty filters, topic-based organization, and structured learning roadmaps.
AI-powered resume screening system with automated keyword extraction, skill matching, and candidate scoring. Built with Streamlit, DVC for data versioning, MLflow for experiment tracking, and comprehensive CI/CD pipeline with automated testing and deployment.
A premium, client-side web application that instantly converts YAML configurations into Python variable assignments with a modern glassmorphism UI. Fully responsive and visually stunning.
Agentic customer support chatbot for a food delivery platform. Uses a LangChain SQL agent to query a live SQLite order database and return real-time, policy-aware responses via Gradio, with input/output guardrails and escalation handling powered by Llama 4 Scout on Groq.
End-to-end medical imaging pipeline classifying pneumonia from chest X-rays using a custom CNN and five transfer learning architectures (VGG16, ResNet50, InceptionV3, EfficientNet, DenseNet). Features DICOM support, Grad-CAM interpretability, class imbalance handling, and a Dockerized Gradio interface.
Full-stack RAG application for document Q&A over PDF, DOCX, and TXT files. Responses are grounded exclusively in uploaded document content via Supabase pgvector similarity search and Cohere embeddings, with a React frontend and FastAPI backend.
Client-side code and text diff tool with three viewing modes — Side-by-Side, Unified, and Character Diff — plus Jupyter notebook support. Built with React and jsdiff, fully browser-based with no backend required.
Contribution history · last 12 months
Activity graph
Repos per language
Commit language
I'm always open to discussing new projects, opportunities, or collaborations. Whether you have a question or just want to say hello, feel free to reach out.