Software Engineer - ML Platform

Overland AI•Seattle, WA

8d•$120,000 - $300,000

About The Position

In this role, you will design and implement Overland AI’s infrastructure to enable training, evaluating, and deploying machine learning/deep learning models for autonomous vehicles. This includes writing software for fast curation and creation of high-quality labels for various robotics tasks using state-of-the-art AI and human feedback. You will also design and implement data processing pipelines, dashboards, and visualization tools for model evaluation and dataset insights, as well as distributed workflows for data processing, training, and evaluation (on-prem and cloud). You will be responsible for on-prem and cloud infrastructure including data storage, compute resources and access, and leverage CI/CD practices to make the machine learning lifecycle efficient and reproducible. You will collaborate closely with experts in Perception and Behavior Planning.

Requirements

Strong programming skills in Python and/or C++, including unit and integration testing
Strong understanding of data processing pipelines for training ML models
Experience with batch processing tools, such as Spark, Beam, Ray, etc.
Experience with data stores and databases, such as MinIO, Postgres, Redshift/BigQuery, etc.
You excel in a small-team atmosphere, taking ownership of problems and working with your colleagues to solve problems across disciplines.
Ability to obtain and maintain a DOD Security Clearance.

Nice To Haves

Prior industry experience in a similar software infrastructure or MLOps team
Prior experience with ROS or other robotics middleware
Contributions to any related open source software project

Responsibilities

Design and implement Overland AI’s infrastructure to enable training, evaluating, and deploying machine learning/deep learning models for autonomous vehicles
Writing software to enable the fast curation and creation of high-quality labels for various robotics tasks using state-of-the-art AI and human feedback
Design and implement data processing pipelines, dashboards, and visualization tools for model evaluation and dataset insights
Design and implement distributed workflows, including data processing, training, and evaluation (on-prem and cloud)
Be responsible for on-prem and cloud infrastructure including data storage, compute resources and access
Leverage CI/CD practices to make the machine learning lifecycle efficient and reproducible
Collaborate closely with experts in Perception and Behavior Planning