AI Scene Lighting Intelligence Developer

Helvex Studio

Remote · $130k – $160k / year · Full-time · Posted 4 months ago

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

This role is no longer accepting applications

Don't miss the next role like this

Create a free profile with your skills. We match you to new AI roles as they open, and companies hiring can contact you directly.

Create free profile, 2 minutes

Similar open roles you can apply to now:

See all open AI jobs

About the role

A VFX lighting TD spends 2 weeks manually matching the lighting of a CG character to the live-action plate it is composited into. The reference is there — it is visible in every frame of the background plate. The challenge is extracting an accurate lighting model from that reference and applying it consistently across hundreds of character renders at different positions and angles in the scene. We build AI tooling for VFX and animation pipelines. Our lighting intelligence system estimates scene illumination from live-action footage and transfers it to CG elements automatically. The current system handles simple one or two-light setups well. Complex outdoor environments with sun, sky, bounce light, and practical sources are where accuracy degrades. You will work on the illumination estimation model: improving accuracy on complex multi-source environments, handling HDR estimation for specular surfaces, and building the uncertainty quantification that tells lighting artists where to apply manual corrections versus where the AI estimate is trustworthy enough to keep.

Fully remote.

Responsibilities

Improve illumination estimation accuracy for complex multi-source outdoor scenes
Build HDR lighting model extraction from standard dynamic range footage
Develop uncertainty quantification for illumination estimates
Design per-frame consistency checks ensuring lighting estimates are temporally stable
Build evaluation benchmarks comparing AI estimates against manually crafted lighting setups
Integrate with standard VFX pipeline formats: EXR, OpenEXR, USD

Requirements

5+ years in computer vision or graphics ML with rendering, VFX, or lighting experience
Experience with intrinsic image decomposition, illumination estimation, or inverse rendering
Strong PyTorch skills and understanding of physically-based rendering concepts
Familiarity with VFX pipeline tools and formats
Portfolio or publications in scene understanding, lighting, or rendering AI