I
Principal Computer Vision Engineer
Not specifiedFull-timeAI & Machine LearningLogistics
Join a small AI team developing computer vision and augmented-reality technology for real-time object detection, scanning, spatial understanding, and retail analytics on mobile devices.
Responsibilities
- Architect and evolve computer vision pipelines from research and training through production deployment.
- Design classic computer vision algorithms and deep learning architectures for spatial understanding, alignment, and tracking.
- Train, evaluate, and fine-tune models for detection, classification, and segmentation.
- Build and optimize low-latency inference pipelines in Python and C++ using CUDA and TensorRT.
- Set technical direction for AI training and inference systems and evaluate new architectures.
- Own technical design, implementation, observability, and iterative improvement of the perception engine.
- Collaborate with mobile and backend engineering teams to deliver production capabilities.
Requirements
- 6+ years of industry experience in computer vision, machine learning, or a related field.
- Strong practical expertise in optical flow, multi-view geometry, feature matching, homography estimation, and OpenCV.
- Strong understanding of deep learning for object detection, image classification, and semantic or instance segmentation.
- Production-level Python programming skills.
- Extensive experience with PyTorch and scalable model training and evaluation pipelines.
- Proven ability to optimize complex algorithms for resource-constrained or real-time deployment.
Nice to have
- Experience with advanced segmentation and foundation models such as SAM, Mask2Former, or Grounding DINO.
- Experience integrating computer vision models into native iOS or Android SDKs.
- Experience translating computer vision algorithms into Apple frameworks and iOS.
- Background in retail technology or augmented-reality scanning applications.