AI Automated Assembly Verification Engineer

Daeva Tech

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

Assembly verification catches errors that defect inspection misses. A PCB can pass all dimensional and solder joint checks but fail in the field because a capacitor was installed in the wrong orientation, a connector was seated at 2 degrees off-axis, or two identical-looking components were swapped. These errors are rare but catastrophic — and they're invisible to the standard vision systems that check for defects rather than identity. You'll build the assembly verification models that cross-check as-assembled state against the bill of materials and assembly instructions. The challenge is multimodal: you need to combine component recognition from RGB imagery, orientation detection from 3D point clouds, and BOM matching from structured product databases to produce a binary pass/fail with a localized failure explanation. Explainability is non-negotiable — assembly technicians need to know exactly which component failed and why. The computer vision team is six engineers. You'll own the verification pipeline, which is distinct from the defect detection pipeline (maintained by another engineer). The two systems share infrastructure but have separate model architectures and evaluation criteria.

Fully remote.

Responsibilities

Build multimodal assembly verification models combining RGB, 3D, and BOM data
Design component recognition and orientation detection for as-assembled state
Implement BOM matching to verify component identity against product specifications
Develop localized failure explanation for assembly errors caught by the model
Create evaluation benchmarks covering component swap, orientation, and placement errors
Optimize inference throughput to support production line cycle times

Requirements

5+ years in computer vision with industrial inspection or assembly verification experience
Experience with multimodal vision architectures combining 2D and 3D inputs
Strong PyTorch skills with production model deployment experience
Understanding of BOM structures and manufacturing assembly processes
Track record building systems with localized, human-readable failure explanations