Senior Machine Learning Engineer - Auto-labeling / Active Learning
About You and The Role
In service of our mission to operate at global scale, we’re growing our perception capabilities, to expand quickly and safely into new products and locations, with the ultimate goal of delivering essential packages right to your doorstep. We are looking for a passionate and creative Perception Software Engineer to join Zipline’s Autonomy Data team. The Autonomy Data team at Zipline is responsible for building high quality ML datasets at scale, used to train ML models that power Ziplines perception centric capabilities on its vehicles.
What You'll Do
- Lead development and set roadmap of our data curation, active learning, and automated labeling strategies for the capabilities we are developing. The core ROI is speed of annotation and quality of labels to enable us to get the best datasets into our system.
- Lead development on open image data exploration, image retrieval, and rule-based/model-based data curation methods
- Identify areas for improvement in existing models, optimize them, and deploy improvements without regression
- Contribute to state-of-the-art machine learning infrastructure and relevant software (e.g. distributed training, continuous model integration, data management, and evaluation of production systems).
- Address large scale challenges in the machine learning development cycle, especially around distributed training in the cloud and data engineering
- Stay up to date on the state-of-the-art in deep learning ideas and software
- Collaborate with the capability teams to implement cutting-edge deep learning modes to accelerate model training time, improving performance, and tackle open problems
- Understand the inner workings of neural networks to uncover edge cases and make safety determinations
- Identify and mitigate bottlenecks in our machine learning development processes
What You'll Bring
- Master or Doctoral degree, more than 3 years of deep learning algorithm research, and project experience with application in computer vision
- Successful application of active learning and auto-labeling into AI development process
- Familiarity with development of data driven MLOps pipeline to iterate models on incoming data
- Experience building reproducible data and machine learning pipelines
- Experience in open data exploration and curation to improve deep learning models.
- Experience working with cloud technology stack (eg. AWS or GCP) and developing machine learning models in a cloud environment.
- Deep understanding of the theory and practice of modern machine learning techniques
- Clear grasp on basic linear algebra, optimization, statistics, and algorithms.
- Experience working with Pytorch, Tensorflow, or other modern deep learning frameworks in a production setting.
- Computer vision experience not required, but recommended.
- Nice to Have: Published research in areas of machine learning at major conferences (NeurIPS, ICML, EMNLP, CVPR, etc.) and/or journals
What Else You Need to Know
The starting cash range for this role is $180,000 - 225,000. Please note that this is a target, starting cash range for a candidate who meets the minimum qualifications for this role. The final cash pay for this role will depend on a variety of factors, including a specific candidate's experience, qualifications, skills, working location, and projected impact. The total compensation package for this role may also include: equity compensation; overtime pay; discretionary annual or performance bonuses; sales incentives; benefits such as medical, dental and vision insurance; paid time off; and more.
Zipline is an equal opportunity employer and prohibits discrimination and harassment of any type without regard to race, color, ancestry, national origin, religion or religious creed, mental or physical disability, medical condition, genetic information, sex (including pregnancy, childbirth, and related medical conditions), sexual orientation, gender identity, gender expression, age, marital status, military or veteran status, citizenship, or other characteristics protected by state, federal or local law or our other policies.
We value diversity at Zipline and welcome applications from those who are traditionally underrepresented in tech. If you like the sound of this position but are not sure if you are the perfect fit, please apply!