Looking for a job

Rohit Deshmukh

@rohitdeshmukh1

Message

I build reliable LLM, computer vision, and self-healing MLOps systems.

India

Message

What I'm looking for

I’m looking for a role where I can work on meaningful AI problems, continuously learn and grow, and contribute to products that have a real-world impact. I value a collaborative environment where I can take ownership, solve challenging problems, and grow with the team.

I've built LLM and agentic systems including a hallucination-controlled RAG pipeline for scientific documents, an autonomous multi-agent data analytics system, and MLOps AutoShield, a self-healing model lifecycle pipeline.

At Central Water & Power Research Station, Pune, I developed a Sentinel-2 image analytics pipeline using a Siamese U-Net to segment four levels of building-damage severity across pre- and post-disaster imagery. I also researched satellite-derived bathymetry methods for estimating water depth from optical imagery.

I work across Python, LangChain, LangGraph, FastAPI, PostgreSQL, Docker, and ML observability tooling. My peer-reviewed work on Temporal Convolutional Networks for predicting RR Lyrae variable-star light curves was accepted at BRAINCON 2026.

Experience

Work history, roles, and key accomplishments

AI/ML Engineer

Central Water & Power Research Station (CWPRS)

Sep 2025 - Jan 2026 (4 months)

Developed an image analytics pipeline on Sentinel-2 multispectral imagery using a Siamese U-Net for pixel-level segmentation of building damage severity. Researched inverse-problem techniques for satellite-derived bathymetry to estimate water body depth from optical bands.

Python Sentinel 2 Siamese U Net Segmentation Satellite Imagery Remote Sensing Inverse Problems Research