Bucket Robotics - (YC S24)

"AI using vision systems can identify anomalies and defects at speeds and accuracies that are not possible with human inspection." – Plastics Decorating Magazine We couldn’t agree more! At Bucket Robotics - (YC S24), we’ve taken this vision and turned it into a reality. Our AI-powered vision systems identify defects that traditional methods miss, enabling manufacturers to catch issues early and improve their processes. No more reliance on labor-intensive human inspection—our systems provide faster, more reliable results. The future of plastics decoration and defect detection is here.

See Bucket Robotics' Vision System in Action! We're showcasing how Bucket Robotics’ defect detection system is revolutionizing part inspection. By transforming a low-cost webcam into a powerful defect detector, our AI-driven software solution proves that it's not just about the hardware—it's about the quality of the software. Our vision system accurately detects defects in parts, commanding automation solutions to sort the good from the bad. Designed for deployment in a wide variety of conditions, our solution makes automation more accessible and scalable for manufacturing. Watch how we’re bringing precision and efficiency to the next level: https://lnkd.in/eKKD75Xk

Meet Bucket’s newest employee – a robotic manipulator! This addition to our team will help us simulate real-world manufacturing environments as we push the boundaries of defect detection. Excited to see what our new "employee" helps us accomplish next! 

Quality Assurance in manufacturing is challenging. How do you tackle it? At Bucket Robotics, we're using synthetic data to create high-quality detection models for manufacturing! Vote in the poll or share your thoughts below. Let’s discuss in the comments—your insights might inspire new solutions!

A port of DINOv2 to JAX has recently appeared in the Hugging Face Transformers repository! JAX has been somewhat limited in off-the-shelf models for vision, which makes this an especially good addition. At Bucket Robotics, we're always exploring the latest advancements to push the boundaries of defect detection in manufacturing. This could open up new possibilities for improving training times -- leading to faster model deployments! Check it out here: https://lnkd.in/d-ZWG7PH

Just got our new PCB NFC business cards from Diode Computers, Inc.! They’re so cool, Diode even made the little LED light up orange (our company color) when the card activates! Hopefully I’ll get a lot of use out of them at #IMTS.

🚀 We launched Bucket Robotics on Hacker News, and the response has been incredible! 🎉 At Bucket, we’re transforming defect detection in manufacturing by creating custom models directly from CAD designs. This approach allows us to deliver ready-to-use detection models before the mold's installed, saving time and reducing errors. Defect detection is critical in the $300 billion injection molding market, and our solution is making it faster, more accurate, and more accessible. We’re thrilled to share this journey with the community and continue pushing the boundaries of what’s possible in manufacturing automation. Check out our post on Hacker News to learn more: https://lnkd.in/g6GWn23r

𝗜𝗻𝗷𝗲𝗰𝘁𝗶𝗼𝗻 𝗠𝗼𝗹𝗱𝗶𝗻𝗴 𝗮𝗻𝗱 𝘁𝗵𝗲 𝗛𝗶𝗱𝗱𝗲𝗻 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 𝗼𝗳 𝗖𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝗰𝘆 Injection molding is a process that requires precision-machined molds—typically made from steel or aluminum. However, even when a manufacturer uses two or more machines to produce the same part, the parts coming out of the molds are never truly identical. This is because each mold is machined slightly differently, leading to unique failure rates and specific zones where defects are likely to occur. For example, one mold might be milled just a fraction of a millimeter lower on one side, which could increase the likelihood of issues like flashing on your part. Traditional defect detection systems rely heavily on samples and labels, requiring a significant amount of data to create a system that works across different production lines. At Bucket Robotics, we tackle this inconsistency head-on. By incorporating these tolerances into our defect detection build process, we use CAD and synthetic data to deliver tailored models rapidly. This approach not only saves time but also enhances the accuracy and reliability of defect detection across multiple lines, ensuring that every part meets the highest quality standards.

One of the most awaited blogs on Bucket Robotics - (YC S24) bringing edge ML to manufacturing industries. Thank you Matt for sharing your insights and thoughts on how the industry is growing and adopting AI applications! Written with: Harini Anand Share this with your network to keep them updated with startup stories building exciting AI tools ♻

We're counting down the days until IMTS - International Manufacturing Technology Show in Chicago! Our CTO Stephan Wolski will be there representing Bucket and talking about computer vision and manufacturing!