warming /agents /rag /mlops
Building scalable, production-ready AI and software systems across Computer Vision, NLP, and Agentic AI.
7+ years • Senior ML Engineer & SWE — production deployments, agentic orchestration, and CI/CD at scale.
Senior Machine Learning Engineer and Software Engineer with over 7 years delivering enterprise-grade AI and software solutions. I focus on moving models into production — writing reliable model code, robust APIs, and repeatable deployment pipelines.
My work spans Computer Vision, NLP, and agentic systems that automate analytics and workflows. I prefer shipping maintainable code and building observability into ML systems from day one.
I enjoy open-source tooling and pragmatic engineering: making complex systems simple to operate and measure.
Joined Google as an FDE, focused on building and delivering production software and AI systems.
Led engineering for Fractal AI (India's First AI Unicorn Company), a multi-agent orchestration platform (LangGraph / LangChain). Owned tool onboarding and metadata lifecycle to productize agent behaviors for largest US telecom's business use.
Implemented human-in-the-loop feedback and automated deployment flows; added Label Studio + FastAPI annotation pipelines, Trivy scanning, MLflow tracking and Seldon serving.
Built a real-time liveliness detection service (MTCNN + FaceNet), converted prototypes to modular MLOps pipelines, and established monitoring + retraining flows on AWS.
Delivered AI + ETL systems for smart-factory operations; shipped a 24/7 IPQC audit system (≈200 hours/week saved) and dashboards across 20+ vendor sites.
Completed internships at Tata Consultancy Services and Infosys, gaining hands-on experience in machine learning, data analysis, and software development practices.
Image classification deployed as a Telegram bot. Integrated Weights & Biases for experiment & artifact tracking and a simple MLOps workflow for continuous improvements.
High-accuracy ALPR system with end-to-end pipeline and a Streamlit web app for demo and deployment.
Anti-spoofing pipeline (MTCNN + FaceNet) optimized for low-latency deployment on AWS with pruning/quantization applied.
Real-time deep learning classifier for X-ray images, with notebooks and deployment tooling for evaluation and demo.
Autonomous data-analysis agents that automate EDA, visualization, and report generation — implemented as a folder inside Data-Science-Projects.
Extracts text from PDFs and converts to podcast-style audio using TTS + summarization, with demo scripts and deployment notes inside the parent repo.
Collection of notebooks and demos covering model prototyping, data pipelines, visualizations and small utilities.
Loading more projects...
I am eager to join your team and contribute my expertise in Agentic GenAI, software engineering and production-grade ML delivery.
Find me at: