Computer Vision Jobs

This role focuses on defining the future of robotics data as a Robotics Researcher. You will work at the forefront of R&D, shaping how embodied data is collected, structured, and used for robot learning and perception. The position involves standardizing data formats, designing innovative collection methods, and collaborating with cross-functional teams to drive breakthroughs in AI training.

This role involves recording short videos of physical activities using a smartphone to collect motion and video data for AI and robotics systems. Contributors follow clear recording guidelines and are expected to submit at least 10 hours of approved video data per week. No advanced technical background is required, but strong attention to detail and consistency are essential. It is a flexible, remote contractor opportunity with fast onboarding.

This role involves training next-generation AI systems by capturing precise motion data using a smartphone's IMU sensor and recording synchronized video footage. You will follow strict technical protocols to ensure high-quality data submissions while working remotely from specific US states. No prior AI experience is required; domain expertise in data collection and attention to detail are key.

This role leverages healthcare expertise to train next-generation AI systems. As a contractor at micro1, you will analyze and annotate medical images and videos, develop challenging questions requiring clinical acumen, and collaborate with a network of experts to ensure question integrity and quality. Your work directly shapes how AI models learn and reason across various healthcare domains.

This role involves applying machine vision expertise to train next-generation AI systems by designing robust algorithms for image analysis and automated inspection. You will use OpenCV and other computer vision libraries to build scalable solutions, optimize image processing pipelines, and collaborate with cross-functional teams. The position focuses on leveraging domain knowledge rather than prior AI experience, making it ideal for experienced machine vision engineers.

This role involves applying hands-on video production expertise—such as editing, filming, VFX, or motion graphics—to train next-generation AI systems. You will analyze video content, provide detailed feedback, annotate datasets, and evaluate AI-generated video outputs to improve model accuracy and quality. No prior AI experience is required; your professional video knowledge is the core qualification.

This role involves reviewing short (5-second) videos of robots performing physical tasks and analyzing corresponding AI-generated captions. The specialist detects discrepancies, hallucinations, or omissions, then rewrites captions for accuracy and clarity to help train next-generation AI systems. It requires strong editing skills and attention to detail, but no prior AI experience is necessary.

As an AI Data Reviewer at micro1, you will help train next-generation AI systems by reviewing and annotating screenshots of software interfaces. Your role involves evaluating AI model responses based solely on visible screen evidence, distinguishing facts from assumptions, and providing clear written justifications. This remote contractor position leverages your domain expertise—no prior AI experience required—to improve model reasoning and performance.

This role involves collecting synchronized video and motion sensor data using a smartphone to train machine learning models for activity recognition and robotics. As a contractor, you will perform physical activities while recording first-person video and IMU data, following strict protocols. You must deliver at least 10 hours of approved video per week, working remotely from specified Latin American and Caribbean countries.

This role involves creating and optimizing parametric 2D and 3D models using OpenSCAD to train next-generation AI systems. The work will help shape how models learn and reason through high-quality real-world input. No prior AI experience is required; domain expertise in OpenSCAD is what matters.