Lead – Physical AI

Location: Bengaluru, IndiaExperience: 3+ yearsRole Type: Full-time

About the Role

We are making a significant investment in Physical AI and are looking for a hands-on technical leader to build and lead this capability from Bengaluru.

The role will focus on applying computer vision, video intelligence, multimodal AI, and machine learning to real-world industrial environments.

We are looking for someone who can move from research to prototype to production, while gradually building and leading a strong Physical AI team.

This is not a conventional engineering-management role. We want a builder who can become a technical leader.

What You Will Work On

You will build AI systems that can understand physical activity from video, cameras, sensors, and other real-world signals.

Key problem areas include

- Video and activity understanding
- Vision-Language and Video-Language Models
- Object, equipment, and tool recognition
- Action and workflow recognition
- Temporal and sequence understanding
- Human-object interaction
- State-change detection
- Workflow deviation and anomaly detection
- Time-and-motion analysis
- Real-time or near-real-time AI assistance
- Edge and cloud-based inference

You will also evaluate emerging technologies across Physical AI, world models, multimodal foundation models, robotics perception, NVIDIA Cosmos/Metropolis, synthetic data, and edge AI.

What You Will Own

You will

- Define the technical direction for the Physical AI capability
- Build and validate prototypes quickly
- Decide when to use foundation models, specialized models, fine-tuning, or traditional computer vision
- Design scalable video and multimodal AI pipelines
- Take successful prototypes into production
- Establish evaluation,



monitoring, and inference architecture
- Work closely with product leadership on real-world use cases
- Recruit and mentor engineers as the capability grows
- Help shape the company’s long-term Physical AI strategy

What We Are Looking For We care more about what you have built than the number of years on your résumé.

You should have 3+ years of relevant experience in areas such as:

- Physical AI
- Computer Vision
- Multimodal AI
- Robotics
- Video Intelligence
- Applied Machine Learning

Strong hands-on experience in several of the following is expected:
- Python and PyTorch
- Deep Learning and Transformers
- Computer Vision
- Video processing
- Vision-Language Models
- Object detection, segmentation, and tracking
- Human-action or activity recognition
- Temporal modelling
- Model training and fine-tuning
- GPU inference
- AI/ML evaluation

Particularly Relevant Experience We are especially interested in candidates who have worked on:

- Industrial computer vision
- Robotics perception
- Autonomous systems
- Manufacturing AI
- Video analytics
- Visual inspection
- Workflow or activity recognition
- Edge AI
- AR / wearable AI
- World models or embodied AI

Direct industrial experience is practical but not mandatory. Educational Background A strong academic foundation in Computer Science, AI, Machine Learning, Computer Vision,



Robotics, Electrical Engineering, or related fields is preferred.

Graduates from IITs and other leading institutions are particularly encouraged to apply.

A Master’s or PhD is valuable but not required. Strong product-building, startup, research, or open-source experience can outweigh academic pedigree.

What Will Make You Stand Out

You may be an exceptional fit if you have:

- Built a real-world computer-vision or multimodal AI product
- Worked extensively with video rather than only images
- Built real-time vision or edge-AI systems
- Fine-tuned multimodal or video models
- Worked in robotics or autonomous systems
- Used NVIDIA’s AI ecosystem
- Contributed to relevant open-source projects
- Published research in areas such as CVPR, ICCV, ECCV, NeurIPS, ICLR, ICRA, IROS, or similar venues
- Built an AI startup or significant independent product

Who We Are Looking For

You are

- Highly hands-on
- Curious about frontier AI
- Comfortable with ambiguity
- Product-oriented
- Pragmatic about technology choices
- Able to move quickly from research to working systems
- Entrepreneurial
- Capable of leading from the front

Why This Role This is an opportunity to build a new AI capability from the ground up.

You will help define the technology, architecture, roadmap, and team — and work on ambitious problems at the intersection of AI, computer vision, multimodal intelligence, and the physical world.

Pay: ₹2,500,000.00 - ₹3,000,000.00 per year

Benefits

- Food provided
- Health insurance
- Paid sick time
- Paid time off
- Provident Fund

Work Location: Hybrid remote in Bengaluru, Karnataka (Bengaluru Urban District)

📌 Lead – Physical AI (Bengaluru)
🏢 VRIO Digital Services
📍 Bengaluru