About Bucket Robotics

At Bucket Robotics, we’re on a mission to revolutionize quality control in manufacturing. Today, over $10 billion worth of manufactured parts are scrapped due to defects, and traditional inspection methods are slow, expensive, and unreliable. We’re changing that by deploying AI-powered vision systems that detect defects faster, more accurately, and with minimal setup.

Our unique approach uses synthetic data generated from CAD models, allowing us to deploy defect detection models 50x faster than competitors—no need for thousands of real-world images.

We’re a team of engineers and innovators with backgrounds from Michelin, Uber ATG, Argo AI, and Stack AV, applying our robotics expertise to solve real-world manufacturing challenges. We’ve already secured enterprise customers in automotive and plastics, and we’re just getting started.

If you're excited about AI, robotics, and making a tangible impact on manufacturing, we'd love to hear from you!

Role Overview

Bucket Robotics is hiring a Machine Learning Engineer to push the frontier of CAD-native computer vision for manufacturing. You’ll work on the core ML systems that turn 3D geometry and synthetic data into reliable, production-grade vision models deployed on factory floors.

This role is perfect for someone who gets genuinely excited about new research ideas, enjoys deep analytical thinking, and wants to see those ideas survive contact with reality—edge deployment, distribution shift, weird lighting, and all.

You’ll operate at the boundary between research and production: exploring new approaches, validating them rigorously, and shipping the ones that work.

Responsibilities

• Design, train, and evaluate computer vision and ML models for inspection and perception
• Develop novel approaches for learning from CAD, synthetic data, and limited real-world data
• Run rigorous experiments to understand model behavior, failure modes, and tradeoffs
• Improve model robustness across lighting, materials, viewpoints, and manufacturing variation
• Optimize models for edge deployment (latency, memory, reliability)
• Collaborate closely with engineering to integrate models into production systems
• Build evaluation frameworks and metrics that reflect real-world manufacturing requirements
• Stay current with cutting-edge research and proactively test promising ideas

Required Qualifications

• 3+ years experience developing ML systems in research or production environments
• Strong foundation in machine learning fundamentals and statistical reasoning
• Hands-on experience training deep learning models (e.g. PyTorch)
• Strong Python skills and comfort working in experimental codebases
• Enjoys deep problem-solving, experimentation, and analytical thinking
• Comfortable operating in ambiguity and iterating quickly

Preferred Qualifications

• Experience in computer vision, 3D vision, or geometry-aware ML
• Familiarity with synthetic data generation, simulation, or domain randomization
• Experience deploying models to edge or resource-constrained environments
• Background in robotics, manufacturing, or physical-world ML systems
• Authored technical reports, blog posts, or peer-reviewed research
• Previous startup or early-stage company experience

What We Offer

• Competitive salary & meaningful equity – help define the core ML systems of the company
• Comprehensive benefits – health, dental, and vision insurance
• Professional development budget – conferences, courses, and research exploration
• A chance to work on hard problems – ML that has to work in the real world, not just benchmarks

Our Values

• Innovation First: We push the boundaries of what's possible in ML and advanced manufacturing
• Ownership Mentality: We take responsibility and act like founders
• Collaborative Spirit: We work together to solve complex challenges
• Continuous Learning: We're always growing and improving

Bucket Robotics is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees.

Tech at Bucket Robotics

At Bucket Robotics, we’re building AI-powered vision systems that solve real manufacturing problems—fast. Our stack is designed for rapid iteration, photorealistic synthetic data generation, and easy deployment to factory floors.

We work across the full stack of industrial AI:

AI & Vision

Custom computer vision models trained entirely on synthetic data generated from CAD—no real-world defects required.

Synthetic Data Generation

Blender-powered rendering pipelines with industrial materials, realistic lighting, and defect modeling, all configurable from the browser.

Edge Deployment

Models exported to ONNX and deployed on NVIDIA Jetson and x86 platforms for real-time inference on the line.

Manufacturing Integration

Outputs connect directly to manufacturing systems—ERP (SAP, Oracle, etc.), MES, and quality tracking—ensuring that inspection insights feed into production workflows.

Developer Experience

We build with Python, TypeScript, and WebGL (Three.js), using a modern stack (React 19, TailwindCSS, Vite, Flask) to maximize speed, usability, and flexibility.

We're solving the cold-start problem in machine vision with an engineer-friendly platform built for real production environments. If you care about closing the loop between CAD, QA, and deployment—let’s talk.