Lead Computer Vision Engineer

Please watch this 2-minute video first:

https://www.loom.com/share/90ad343b73884d819d9c4e8aeba5966c

Self Inspection | Remote — Latin America | 7+ years experience

About the Role

Self Inspection enables automated vehicle inspections using computer vision and machine learning. Users capture photos and videos of vehicle exteriors, interiors, wheels, tires, and damage — analyzed by our AI systems and experts to generate detailed condition reports. Over one million inspections completed, customers among the largest automotive companies in North America, backed by top Silicon Valley VCs.

As Lead Computer Vision Engineer, you'll own the end-to-end computer vision roadmap — from data collection and model development to deployment, monitoring, and continuous improvement in production. You'll work closely with executives, product, and engineering to build scalable AI that directly impacts customers.

This is a hands-on leadership role: you'll write code, train models, review architectures, mentor engineers, and make the key technical decisions. Great fit if you want to set the technical direction of a product where computer vision IS the product. Not a fit if "lead" means you stopped coding.

What You'll Do

- Define and execute the computer vision strategy and roadmap; drive decisions on architectures, infrastructure, tooling, and data pipelines
- Design, train, evaluate, and deploy models for object detection, instance/semantic segmentation, classification, tracking, OCR, image quality assessment, and 3D vision
- Build automotive-specific solutions: damage detection, vehicle part detection, tire analysis, condition assessment, document and VIN recognition
- Design data collection, labeling, and active learning workflows; build scalable training pipelines with versioning, experiment tracking, and reproducibility
- Deploy to cloud and edge — optimizing inference with ONNX, TensorRT, TorchScript, or OpenVINO
- Establish production monitoring for model drift, data quality,



and performance
- Mentor computer vision and ML engineers; set best practices for development, evaluation, and deployment
- Translate business requirements into AI solutions and communicate tradeoffs to stakeholders

Success looks like: higher production model accuracy, fewer false positives and negatives in damage detection, scalable AI infrastructure for rapid experimentation, clear evaluation and monitoring standards, and a growing, well-mentored CV team. Must-Haves
- 7+ years of hands-on Computer Vision and Machine Learning experience
- 2+ years leading or mentoring engineers and driving technical direction
- Proven track record deploying CV systems to production at scale — owning models beyond training, through deployment, monitoring, and maintenance
- End-to-end ML pipeline design, from data acquisition to production inference
- Expert Python and PyTorch; strong OpenCV
- Deep experience with modern detection/segmentation architectures: YOLO (v8–v11), RT-DETR, Faster/Mask R-CNN, Vision Transformers, foundation vision models
- Inference optimization: ONNX, TensorRT, or TorchScript; quantization and profiling
- MLOps: MLflow or Weights & Biases, Docker, Kubernetes, CI/CD; AWS, GCP, or Azure
- Strong English — you'll work directly with executives and the U.S. team
- Located in Latin America with 4 hours of daily overlap with PST (7–11 AM)

Nice-to-Have Automotive CV applications · damage detection, vehicle inspection, or insurance AI · real-time mobile vision · 3D vision, depth estimation, photogrammetry, Gaussian Splatting · multimodal AI combining images, video, and text · on-device deployment (iOS/Android) · MMDetection, Albumentations · publications, patents, or open-source contributions

We don't expect all of it. Strong on production CV at scale, PyTorch, deployment optimization, and technical leadership — apply.

Hiring Process

1. Apply — you'll be invited to a 20-minute AI interview
2. Latin Remote Workers interview — virtual, with our team
3. Final interviews — with the Self Inspection team

📌 Computer Vision Engineer (Argentina)
🏢 Latin Remote Workers
📍 Argentina