What are the responsibilities and job description for the Labeling Operations Manager position at Stack AV?
About the Role:
The Labeling Operations team is responsible for ensuring a high throughput of quality labels of sensor data used in the development of our Self Driving System. As the Labeling Operations Manager you will manage a metrics driven internal team as well as external vendors responsible for delivering the labels needed by internal customer teams who rely on this data for the development of critical ML models used in Autonomy. You will also partner closely with Product and Engineering teams to identify and define requirements for efficient tooling and automation to streamline labeling and QA processes. This is a critical role to the development and validation of our Autonomy System and will involve close partnership and collaboration broadly across Stack to serve the needs of the organization.
Responsibilities:
- Manage the Labeling Operations team in the Product Development organization
- Collaborate with key customers across the organization to define and continually refine labeling ontology and policy
- Ensure efficient workflows and policies are in place that result in the delivery of a high volume of quality labels meeting customer needs on a continuous basis
- Oversee all aspects vendor relationships and partner organizations responsible for ground truth labeling of AV sensor data, including training, and policy
- Utilize data and analysis to identify opportunities for improvement in operational workflows and processes
Qualifications:
- 10 years of Operations Management or related experience
- Significant experience leading teams responsible for data annotation or related functions
- Demonstrated experience managing globally distributed workforce vendors
- Proven ability to define and implement complex operational workflows that are heavily data driven
- Strong data analysis skills including experience building and reporting on meaningful business metrics
- Prior experience working in an AV or Robotics company is strongly preferred
- General understanding of ML training of Perception models and associated inputs (e.g. calibration), data infrastructure, and visualization tooling preferred
#LI-TT1