Senior Machine Learning Engineer - Perception

Job Description

Job Description

Pickle is on the hunt for a dynamic and driven Senior Machine Learning Engineer to revolutionize the future of warehouse automation. This role will focus on advancing our perception stack by refining our data engines, architecting robust 2D/3D vision models, and deploying them to production hardware to ensure our robots handle the chaotic reality of the physical world.

• Architect Multi-Modal Vision Systems: You will design and train models that fuse 2D inputs with 3D geometry to solve complex grasping and scene understanding tasks.
• Lead End-to-End Model Deployment: You will own the transition from research to reality. This includes model graph optimization, quantization (TensorRT), and runtime integration to ensure low-latency inference on our edge compute hardware (NVIDIA Orin).
• Drive Technical Excellence: As a senior member of the team, you will conduct rigorous code reviews, mentor junior engineers, and contribute to the strategic perception roadmap.
• Own the Data Strategy: You will take ownership of our existing labeled dataset and pipeline. You will identify bottlenecks, improve data quality, implement active learning strategies to systematically resolve edge cases and improve model robustness.
• Ensure Production Reliability: You will write high-performance production code (Python/C++) to seamlessly integrate perception outputs into the broader robotic control stack, prioritizing safety and system stability.

• 5+ Years of Experience in Computer Vision and Machine Learning, with a track record of shipping ML products to the physical world (Robotics, AV, or IoT).
• Expert-level Python and PyTorch skills. Working knowledge of C++ for deployment and system integration.
• Experience with 2D Vision (YOLO, MaskRCNN, Transformers) and 3D Vision (PointNet, grasp generation, multi-view geometry, camera calibration).
• You are proficient with inference optimization tools such as TensorRT, ONNX Runtime, or CUDA to maximize hardware utilization.
• You have experience curating large-scale datasets, detecting statistical bias, and automating quality assurance within the ML pipeline.
• You can translate high-level product requirements into specific engineering tasks and explain technical trade-offs to non-expert stakeholders.
• Familiarity with Docker, AWS/GCP (S3, EC2), labeling platforms and experiment tracking tools.

Pay at Pickle

At Pickle Robot Company, we believe transparency builds trust. The salary range listed here is provided in accordance with Massachusetts law and reflects what we reasonably and in good faith expect to offer for this role. We often consider candidates at different levels of seniority, and final compensation will reflect the level at which a candidate is hired, along with factors like experience and location.

About Pickle Robot

Want to get in on the ground floor of a fast-growing, VC-backed robotics company? Join Pickle Robot! We build systems that companies and their teams love.

Pickle robots unload trucks. Every day, millions of trucks and containers are loaded and unloaded, often requiring manual labor—tough, dirty, dangerous, and hard to staff. Pickle automates this process using AI, machine learning, and robotics to deliver reliable products. Our Unload Systems work with teams on loading docks to make the job safer, faster, and more efficient.

Pickle provides best-in-class benefits including health, dental, & vision insurance; unlimited vacation, along with all federal and state holidays; 401K contributions of 5% your salary, travel supplies, and other items to make your working life more fun, comfortable, and productive.

We may use artificial intelligence (AI) tools to support parts of the hiring process, such as reviewing applications, analyzing resumes, or assessing responses and identifying potential inconsistencies or verification signals in application materials based on available information. These tools assist our recruitment team but do not replace human judgment. Final hiring decisions are ultimately made by humans. If you would like more information about how your data is processed, please contact us.