Pickle Robot Company is looking to solve the problem of rare events in the supply chain, such as novel packaging, damaged goods, and occlusion, by developing strategies to detect and handle these events using robotic perception and machine learning.
Requirements
- 5+ Years of Experience in Computer Vision and Deep Learning, with a specific focus on applied research or advanced R&D.
- You have hands-on experience with modern architectures: Transformers (ViT), Self-Supervised Learning (SSL), Vision-Language Models (VLMs), or Zero-Shot detection.
- Expert-level Python and PyTorch skills.
- Experience with Model Distillation (Student-Teacher networks) and Quantization.
- Familiarity with Docker, AWS/GCP (S3, EC2) and experiment tracking tools.
Responsibilities
- Tame the Long Tail: You will develop strategies to detect and handle 'rare events' (novel packaging, damaged goods, occlusion) that break traditional detectors.
- Leverage Foundation Models: You will adapt and fine-tune large-scale foundation models for specific robotic manipulation tasks, toward zero-shot generalization capabilities.
- Model Optimization: You will take massive, compute-heavy models and distill them into efficient, lightweight networks that can run with low latency on edge compute (NVIDIA Orin).
- Drive Technical Excellence: As a senior member of the team, you will stay ahead of the curve on CV/ML research, translate relevant breakthroughs into production-grade gestures and mentor junior engineers.
Other
- 5+ Years of Experience
- 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.
- Health, dental, & vision insurance
