Loading organizations...

Synthetic data for better vision.
Key people at SBX Robotics.
SBX Robotics was founded in 2020 by Artem Avdacev (Founder) and Joshua Kuntz (Founder) and Ian Dewancker (Founder).
SBX Robotics generates synthetic data that teaches robots to see. We use simulation software to create training data 10x faster and cheaper than annotation services or in-house teams.
Instead of being blocked on data, our clients send 25 images from their robot’s camera, and receive 25,000 perfectly labeled synthetic training images. SBX data is ready to be used by deep learning computer vision models.
Key people at SBX Robotics.
SBX Robotics was founded in 2020 by Artem Avdacev (Founder) and Joshua Kuntz (Founder) and Ian Dewancker (Founder).
SBX Robotics is a Canadian startup specializing in generating synthetic data to improve robotic vision. Their platform uses simulation software to create large, realistic, and perfectly annotated datasets from a small set of real-world images, enabling faster and cheaper training of computer vision models for robots. This synthetic data approach accelerates development by providing 10x faster and more cost-effective training data compared to traditional annotation or in-house data collection. SBX Robotics serves robotics developers and companies needing robust vision capabilities for autonomous systems, drones, and other AI-powered machines, addressing the challenge of acquiring high-quality labeled data for machine learning models. The company has demonstrated growth momentum through its innovative technology and integration with robotic operating systems, supporting rapid iteration and adaptation to changing environments[2][4][6].
SBX Robotics was founded in 2020 by Artem Avdacev and Joshua Kuntz, both with strong backgrounds in engineering and computer vision. Artem previously worked at Yelp on fraud detection and has a passion for computer graphics and robotics, while Joshua has experience from the University of Waterloo and tech companies like Wish. The idea emerged from the intersection of computer vision, robotics, and simulation technology, aiming to solve the bottleneck of acquiring labeled training data for robot vision systems. Early traction included acceptance into Y Combinator’s Winter 2021 batch and developing a platform that converts a small number of real robot camera images into tens of thousands of synthetic, labeled images, enabling clients to bypass traditional data annotation challenges[2][4].
SBX Robotics rides the growing trend of synthetic data and simulation-driven AI training, which addresses a critical bottleneck in robotics and autonomous systems: the scarcity and high cost of annotated real-world data. As robotics applications expand in industries like logistics, manufacturing, and autonomous vehicles, the need for scalable, adaptable vision models grows. Synthetic data allows for rapid iteration and robustness against environmental variability, which is crucial for deploying AI in dynamic real-world settings. The timing is favorable due to advances in deep learning, computer vision, and simulation technologies, alongside increasing demand for automation and AI-driven robotics. SBX Robotics contributes to the broader ecosystem by enabling faster development cycles and lowering barriers for robotics companies to implement advanced vision capabilities[2][4][5].
Looking ahead, SBX Robotics is well-positioned to capitalize on the expanding robotics market and the increasing reliance on synthetic data for AI training. Future trends shaping their journey include improvements in simulation fidelity, real-time adaptive learning, and broader adoption of AI in autonomous systems. Their influence may grow as more robotics developers seek cost-effective, scalable solutions for vision training data, potentially expanding into adjacent markets such as drones, autonomous vehicles, and industrial automation. Continued innovation in their platform’s ease of use and integration will be key to maintaining competitive advantage and driving adoption. SBX Robotics exemplifies how synthetic data can transform AI training workflows, accelerating the path to smarter, more capable robots[2][4][5].