D
Computer Vision Engineer — Structural Inspection
David Mensah
Full-time · Mid-level · New_York
About the role
We are building a robotic system for automated structural inspection of bridges and tunnels. Our sensor package combines LiDAR, structured light, and high-resolution cameras. The vision pipeline processes data from all three modalities to detect surface cracks, spalling, and deformation — defects that currently require a person to abseil down a bridge or crawl through a tunnel with a flashlight and a clipboard. We are pre-revenue. We have completed inspections on four structures under contract with two state DOTs as proof-of-concept engagements. We have $2.1M in grant funding from ARPA-E and a Series A process that we expect to close in Q3. We are telling you this because you deserve to know the risk profile of the company you're joining. This is early, there is risk, and the equity reflects that. What we have is a genuinely hard vision problem, a physical system that works well enough to inspect real infrastructure, and a small team of people who are serious about solving it properly. The vision work involves multi-modal fusion, 3D point cloud processing, and crack detection on textured concrete surfaces — not a clean benchmark dataset.
Responsibilities
- Develop and improve crack and spalling detection models for multi-modal sensor data from bridge and tunnel inspections
- Design data augmentation strategies for structural defect datasets where labelled examples are scarce
- Build evaluation frameworks appropriate to safety-critical defect detection — not just mAP, but precision at relevant recall thresholds
- Contribute to the sensor data processing pipeline from raw capture to model-ready input
- Work alongside civil engineers who understand the failure modes we're trying to detect
Requirements
- 4+ years of computer vision engineering with production or near-production deployment experience
- PyTorch for model development and fine-tuning on small, carefully curated domain-specific datasets
- Experience with 3D point cloud processing — PCL, Open3D, or equivalent — for structural geometry analysis
- Keras or an equivalent framework for rapid prototyping of fusion architectures
- NumPy for the numerical processing that sits between raw sensor data and model inputs
- Experience with multi-modal data fusion is a significant advantage — you've combined signals from different sensor types
- Comfort working with small, expensive, hand-labelled datasets where every annotation matters
Benefits
- Work on a physically real problem with safety consequences — bridges fail, people die, and better inspection matters
- Full remote with access to our lab in Pittsburgh for sensor testing and data collection sessions (travel covered)
- $95,000 – $118,000 base salary + early-stage equity
- Pre-Series A risk — we are transparent about this and the equity is priced accordingly
- $1,500 equipment and tooling budget
Job Type
Full-time
Level
Mid-level
Language
English
Salary Range
$95,000 – $118,000
AI Expertise
AI & Machine Learning Engineers
Ready to apply for this role?
Create a free talent account in under 2 minutes.
- Apply to verified AI companies
- Get AI-matched job recommendations
- Message hiring managers directly
- Build your public AI talent profile