About the client -
We turn plain room photos into realistic “view-in-room” experiences using depth, segmentation, and rendering magic. The stack is fast, modern, and shipping to global brands right now.
What you’ll do
- Own and evolve our CV + genAI pipeline — detection, segmentation, depth, alignment, lighting.
- Build production-ready models and inference APIs that scale cleanly on AWS (GPU Lambda / ECR).
- Push visual realism : correct floor planes, scale, shadows, occlusion, tone.
- Research, fine-tune, and deploy state-of-the-art models
- Collaborate with product and SDK teams to make the tech invisible and fast.
- Shape the technical roadmap and mentor the next hires.
What you’ve done
5+ yrs hands-on in computer vision or applied deep learning.Shipped real systems, not just notebooks.Expert in Python, PyTorch, OpenCV, Docker, and deploying models to production.Comfortable designing modular APIs and optimizing for both latency and realism.Solid grasp of 3D geometry, indoor scene understanding, and rendering principles.Bonus : exposure to Three.js / WebGL / Unreal / Blender pipelines.Why this role
Full ownership of the vision (pun intended).You’ll define how thousands of users see and interact with products.Early equity in a venture-backed company with real pilots and momentum.Work with founders who care deeply about design, speed, and craft.Interested candidates can send resume at shubhamn@usil.tech or can fill the google form - https : / / forms.gle / f92RmdU9mooQ335D9