Founding ML Engineer, Computer Vision (Object Detection)

Accountabilities: Define the computer vision architecture for fine-grained item identification, leveraging foundation models and adapting them to specialized product catalogs. Design, train, fine-tune, and productionize computer vision models capable of accurately identifying items from images. Establish accuracy standards across different product categories and develop calibrated confidence scores that allow systems to recognize and communicate uncertainty. Build evaluation frameworks that measure real-world model performance and track improvements over time. Develop feedback loops that use model errors and production insights to guide labeling, dataset improvements, and future model iterations. Prioritize engineering and research efforts between expanding category coverage and improving accuracy within existing categories. Own the complete path from ML research and experimentation to production deployment, including model-serving latency, cost, scalability, and reliability. Collaborate with technical and non-technical stakeholders to communicate model capabilities, limitations, performance, and tradeoffs. Establish strong foundations for the computer vision function and provide senior technical leadership on an open-ended ML problem. Requirements: 5+ years of applied computer vision experience, including experience shipping a computer vision system to production at meaningful scale. Strong experience with fine-grained or instance-level classification, particularly in scenarios where distinguishing visually similar items is important. Advanced proficiency with PyTorch or TensorFlow and hands-on production experience fine-tuning and deploying vision transformers or convolutional neural networks (CNNs). Experience developing robust evaluation frameworks for measuring real-world model performance, accuracy, and improvement. Strong understanding of the end-to-end machine learning lifecycle, including experimentation, model training, evaluation, deployment, and production monitoring. Ability to operate as the senior technical owner of an ambiguous problem without an established playbook. Strong communication skills and the ability to explain technical concepts, model limitations, and engineering tradeoffs to non-technical stakeholders. Experience with active learning, human-in-the-loop labeling, low-latency model APIs, or early-stage startup environments is a strong plus. Demonstrated curiosity, autonomy, and willingness to take ownership of both strategic technical decisions and hands-on implementation. Benefits: Annual salary range of $200,000–$260,000 . Fully remote work within North America . Founding-level ownership over the computer vision and machine learning function. Significant influence over technical architecture, ML strategy, and engineering priorities. Opportunity to take models from research through production and see their direct impact on a real-world marketplace. High-autonomy environment with the opportunity to solve challenging computer vision problems at an early-stage company.