Loading...
Loading...
Help build the future of AI card grading. Work on CGI Vision AI and push the boundaries of fine-grained image analysis.
As a Computer Vision Engineer at CardGrade.io, you will work directly on CGI Vision AI, the core technology that powers our card grading platform. You will design, train, and deploy computer vision models that analyze trading cards across 47 inspection points, achieving and exceeding our current 92.8% accuracy benchmark.
This is a unique opportunity to work on a fascinating fine-grained visual recognition problem where subtle differences in centering, surface condition, corner wear, and edge quality determine a card's grade and value. You will have access to large datasets of professionally graded cards and the freedom to experiment with cutting-edge approaches.
You will collaborate closely with our grading standards team (led by a former professional grader with 15+ years of experience) to ensure our AI predictions align with real-world grading outcomes.
Develop and improve computer vision models for card grading across all major card types and eras
Train models on large datasets of professionally graded cards to improve prediction accuracy beyond 92.8%
Build features for edge detection, surface analysis, centering measurement, and corner wear classification
Optimize model inference for real-time grading with sub-30-second turnaround times
Collaborate with the grading standards team to align AI predictions with professional grading criteria
Design and run A/B experiments to validate model improvements against real-world grading outcomes
Build data pipelines for ingesting and labeling new training data from card submissions
3+ years of professional experience in computer vision or deep learning
Proficiency in Python and at least one major ML framework (PyTorch or TensorFlow)
Experience with image classification, object detection, or image segmentation
Strong understanding of CNNs, vision transformers, and modern architectures
Experience with model optimization, quantization, and deployment to production
Familiarity with data augmentation strategies and training pipeline design
Strong problem-solving skills and ability to work independently in a remote team
Trading card collecting hobby or familiarity with card grading standards
Experience with fine-grained visual recognition or defect detection
Published research in computer vision or related fields
Experience with edge computing or mobile model deployment (ONNX, TensorRT, CoreML)
Background in image quality assessment or anomaly detection
Meaningful equity stake in a fast-growing company, plus salary that reflects your experience.
Work from anywhere. Our team spans LA, Toronto, London, and Singapore.
Work on state-of-the-art computer vision solving real-world fine-grained analysis problems.
Your work directly impacts 32,000+ cards analyzed and 540+ teams relying on our platform.
Send your resume and portfolio (GitHub, publications, or relevant project links) to:
In your email, tell us about a computer vision project you are proud of and why you are excited about working on AI card grading. We review every application and respond within one week.