About the Role
The role involves developing robust computer vision models, integrating them into production environments, and solving complex perception challenges using real-world data.
Responsibilities
- Design and train deep learning models for visual recognition tasks
- Optimize computer vision pipelines for performance and accuracy
- Work with real-world image and video datasets to improve model generalization
- Collaborate with software engineers to deploy models into scalable systems
- Evaluate model performance using quantitative metrics and real-world testing
- Debug and resolve issues in data labeling, training, and inference
- Stay current with advancements in computer vision and deep learning
- Improve data curation processes to enhance training set quality
- Develop tools for automated annotation and data quality assessment
- Contribute to architectural decisions for end-to-end vision systems
- Write clean, maintainable, and well-documented code
- Participate in code reviews and technical design discussions
- Troubleshoot production issues related to vision model outputs
- Assist in defining requirements for new visual perception capabilities
- Work cross-functionally with product and research teams
- Ensure models meet privacy and ethical standards
- Optimize inference speed and resource usage on target hardware
- Implement testing frameworks for vision components
- Support integration of models into client-facing applications
- Conduct experiments to validate model improvements
- Document technical approaches and system behavior
- Contribute to versioning and reproducibility of model training
- Analyze failure cases to guide future development
- Help define evaluation protocols for new features
- Mentor junior engineers on computer vision best practices
Nice to Have
- Master’s or PhD in computer vision, machine learning, or related field
- Experience with real-time video processing systems
- Background in robotics or autonomous systems
- Published work in computer vision or machine learning conferences
- Experience with edge computing or embedded vision platforms
- Knowledge of 3D reconstruction or multi-view geometry
- Familiarity with active learning or self-supervised learning methods
- Experience optimizing models for mobile or low-power devices
- Exposure to synthetic data generation for training
- Background in human pose estimation or action recognition
Compensation
Competitive salary and equity package
Work Arrangement
Remote - US only
Team
Small, fast-moving team focused on building cutting-edge visual technology
About the Role
This position is central to advancing the core technology by developing accurate and efficient computer vision systems. The engineer will work on full-cycle development from prototyping to deployment.
What We Value
We prioritize practical problem-solving, technical rigor, and the ability to ship reliable systems. Candidates should be comfortable balancing innovation with engineering discipline.
Does not sponsor visas for this position
