Plato Systems

About

We are a series-A startup building perception systems for autonomy. We are based in the San Francisco Bay Area, funded by NEA, and our core team includes faculty entrepreneurs (Stanford and UC Santa Barbara) and industry veterans (Uber, Apple, Amazon Lab126, Rohde & Schwarz), who have successfully shepherded signal processing and machine learning innovations to large-scale software for location improvement and safety at Uber, led the development of state-of-the-art computer vision technologies that shipped over millions of Amazon devices, and delivered zero-to-one product experiences at Uber and Box. Our core product grew out of 5+ years of university R&D by our co-founders. You can find out more about us by visiting our website.

Our mission and team expertise spans beyond software to advanced sensor systems, algorithms, embedded systems, signal processing, and machine learning. Our team is building and deploying edge software and cloud services for real-time customer facing products as well as internal big data tools. We look for people with a depth of expertise and experience in one of these areas, and with the intellectual curiosity for interacting with, learning from, and teaching world-class experts in areas outside their expertise.

We currently have multiple full-time opening in the areas of computer vision and machine learning. The candidates will join a multi-disciplinary team of scientists and engineers and work on a full stack of developing cutting edge Computer Vision (CV) and Machine Learning (ML) methods based on data from a variety of sensors. This position is open to both on-site and remote candidates (including Canada).

Responsibilities

  • Research, design, develop and evaluate advanced image processing and computer vision algorithms for a real-time computer vision pipeline including but not limited to camera calibration, multi-object tracking, object detection and classification, segmentation, and multi-sensor fusion
  • Maintain and improve our existing in-house algorithms and models, including continuous evaluation, gap analysis, re-training and fine tuning.
  • Develop state of the art deep learning networks and architectures across data from multiple sensors; Tasks include training, evaluating, benchmarking and deployment into real-time pipelines
  • Optimize algorithm performance across a wide range of development platforms and embedded systems
  • Develop evaluation scripts to process large data and accurately measure algorithmic and end to end performance.

Basic Qualifications

  • PhD in CV/ML with 1+ years of industry experience or MS in CV/ML with 2+ years of industry experience
  • Strong Python/C++ programming, familiarity with software development best practices, debugging/profiling
  • Understanding of stereo / multi view geometric computer vision and classical computer vision for natural scene images
  • Hands-on experience with OpenCV, PIL, and other image processing libraries
  • Hands-on experience with at least one main stream deep learning framework such as PyTorch, TensorFlow, and ONNX
  • Experience with writing production level code
  • Familiarity with data science toolkit such as jupyter lab/notebooks, pandas, bash scripting, Linux environment
  • Self motivated
  • Excellent problem solving skills
  • Excellent communication skills

Preferred Qualifications

  • Prior experience with multi-sensor calibration and multi-view geometry
  • Hands-on experience with different neural network architectures (CNNs, RNNs, etc.) as well as specific approaches for classification, segmentation, and object detection (Mask-RCNN, SSDs, EfficientDet, …) and common datasets (CoCo, Kitti, nuScenes,...)
  • Solid software engineering foundation and a commitment to writing clean, well architected code
  • Familiarity with various physical aspects of sensors including cameras and Lidars
  • Publications in major CV/ML conferences and journals
  • Statistical modeling, analysis, and significance testing
  • Experience with edge computing (NVidia Jetson family, Raspberry Pi, ML accelerators) and coding for resource-constrained compute environments