The Computer Vision Engineer track covers CV foundations, model architectures (YOLO, ViT, SAM, DINO), detection and segmentation pipelines, annotation and data quality, inference optimization, and production deployment. Sessions probe both technical depth and practical production experience.
The Computer Vision Engineer track covers CNN and ViT architecture trade-offs, object detection frameworks (YOLOv8, DETR), instance and semantic segmentation, dataset annotation and quality, transfer learning, model quantization and TensorRT optimization, and behavioral questions on shipping CV systems to production.
If you describe an object detection pipeline, Alex follows up on your NMS threshold choices, how you handle class imbalance, or how you monitor for distribution shift in production. If you describe a ViT-based approach, Alex asks about your latency trade-offs versus a YOLO-family model.
Yes. The Computer Vision Engineer track includes modern foundation model coverage — SAM for promptable segmentation, DINOv2 for self-supervised pretraining, and CLIP for zero-shot tasks. Sessions cover when to fine-tune a foundation model versus building a custom architecture.
Voice-first, fully dynamic, calibrated to your target level and company.
Practice Computer Vision Engineer interviews →Free session included — no credit card required