Senior Applied Scientist

Samsara Samsara · Enterprise · San Francisco, CA · Remote · Safety AI

Senior Applied Scientist role focused on building and improving ML/CV models for industrial applications, working with large-scale data, optimizing models for inference on backend and edge devices, and collaborating with firmware and full-stack teams for deployment. Requires strong background in Computer Vision and ML tools.

What you'd actually do

  1. Build and improve the accuracy of ML / CV models, including retraining and optimizing models to solve Samsara-specific problems.
  2. Work with petabyte-scale data from Samsara camera and sensor devices to develop new models.
  3. Optimize models for inference on the backend and/or on edge devices.
  4. Partner with firmware and full-stack teams to deploy model for optimal performance and cost.
  5. Stay connected to industry and academic research and adopt novel technology that suits Samsara’s needs.

Skills

Required

  • BS or MS in Computer Science or other technical degree with 5+ years of experience as an Applied Scientist, Machine Learning Engineer, or similar role working on Computer Vision; or Ph.D. in Computer Science or quantitative discipline (e.g., Applied Math, Physics, Statistics) with 3+ years of experience working on Computer Vision.
  • Strong proficiency in one or more common languages (e.g., C++, Golang, Java, Python).
  • Proficiency with common ML tools (e.g., Spark, TensorFlow, PyTorch).
  • Familiarity with managing data processing and machine learning code via GitHub.
  • Excellent problem-solving skills and the ability to work alone and collaboratively.

Nice to have

  • Experience implementing ML models on large datasets 100K to 1M and more on the edge.
  • Experience building, deploying, and optimizing ML models on the edge.
  • Experience in state-of-the-art models for road segmentation, road object detection and tracking.
  • Experience working with cross-functional teams on a project.
  • Comfortable with full-stack/backend development code to build a strong understanding of underlying data structures and other dependencies.

What the JD emphasized

  • ML / CV models
  • retraining and optimizing models
  • petabyte-scale data
  • optimize models for inference on the backend and/or on edge devices
  • deploy model for optimal performance and cost
  • Computer Vision

Other signals

  • build and improve accuracy of ML/CV models
  • retraining and optimizing models
  • petabyte-scale data
  • optimize models for inference on backend and/or edge devices
  • deploy model for optimal performance and cost