Open to opportunities

Nithin Varghese

@nithinvarghese

Message

Senior AI Engineer building production computer vision and generative AI with multi-agent LLM systems.

India

Message

What I'm looking for

I’m looking for a role where I can build generative AI and computer vision products end to end—owning design, ML engineering, deployment, and operations—especially edge-first, real-time systems and multi-agent LLM workflows.

I’m an AI Engineer with 7+ years of experience building and deploying production computer vision and machine learning systems at scale, including 3+ years leading engineering teams as Head of Computer Vision.

I’ve delivered video analytics platforms across mining, energy, banking, and retail—scaling to 10,000+ cameras—and I’m known for hardware-aware model optimization for edge-first, real-time inference using TensorRT, OpenVINO, and TFLite.

Now I’m building generative AI products end to end as founder of an AI product studio—creating multi-agent LLM systems, RAG pipelines, and LLM fine-tuning—while shipping products like LatentStory and FlowWriter.

Experience

Work history, roles, and key accomplishments

Current

Founder & AI Engineer

Current

Vnm Technologies Llp

Dec 2024 - Present (1 year 7 months)

Founded an AI product studio and owned the full lifecycle for generative AI applications, from product design through ML engineering, deployment, and operations. Built a multi-agent storybook generation system using LangGraph and a voice-first ideation tool using LLM agents.

Multi Agent Systems LangGraph RAG Retrieval Augmented Generation AI Agents Prompt Engineering Hugging Face Vector Databases

Head of Computer Vision

Deepsight Ai Labs

Feb 2021 - May 2024 (3 years 3 months)

Led end-to-end delivery of multi-site computer vision systems across mining, energy, banking, and retail, including scoping, architecture, model operations, and on-site rollout. Scaled real-time video analytics and deployed safety- and security-critical use cases such as PPE compliance, intrusion detection, and person-under-steel roll safety detection.

Computer Vision Video Analytics Object Detection Intrusion Detection Facial Recognition TensorRT OpenVINO Edge Computing Raspberry Pi Model Compression Scheduling Monitoring

Computer Vision Engineer

Deepsight Ai Labs

Sep 2018 - Feb 2021 (2 years 5 months)

Created and deployed a camera tamper detection algorithm for production surveillance systems. Trained and deployed deep learning models for masked-face and helmet detection and for crowd detection in safety and security analytics.

Masked Face Detection Crowd Detection Deep Learning

Machine Learning Engineer (Freelance)

Self Employed

May 2018 - Aug 2018 (3 months)

Built a receipt information extraction system that used YOLOv3 for field detection and an OCR pipeline to output structured CSV data from receipt images.

YOLOv3 Receipt Information Extraction Data Extraction CSV