Software Engineer - Perception MLOps
Zipline
About Zipline
About You and The Role
Zipline is on a mission to revolutionize global access to essential supplies through our autonomous drone delivery network. We’re working hard to launch our next generation of autonomous aircraft, known as Platform 2, designed to provide a magical delivery experience for every package we place for a customer.
We need to deliver to the incredibly diverse and challenging geometries of backyards and porches, while navigating a variety of objects we encounter. From complex architectures to swaying trees, from tiny porches to a myriad of kid toys, our challenge is to build a robust autonomy system that puts the package down in a delightfully convenient and safe location every time.
To support this effort, we are seeking a skilled ML Ops Software Engineer who will be instrumental in developing, maintaining and optimizing the tools and technologies that ensure the machine learning models in our perception systems are reliably and continuously improving with data. This role is crucial for maintaining the highest standards of performance and reliability in our autonomous systems, enabling them to adapt and thrive in diverse environments. Your work will ensure that our algorithms perform at their best, leveraging advanced techniques to validate and refine them in real-world conditions.
What You'll Do
- Work closely with our machine learning and data platform engineers to streamline the entire machine learning pipeline.
- Develop and maintain robust frameworks for data curation, data processing and annotation workflows.
- Develop tools for monitoring, validation, and lifecycle management of ML datasets and annotations, and drive adoption of these tools across teams to empower them.
- Establish and manage infrastructure for the continuous integration and continuous deployment (CI/CD) of ML models.
What You'll Bring
- Experience deploying and maintaining data pipelines for machine learning systems in production environments.
- Strong familiarity with cloud infrastructure like AWS or GCP, and distributed systems for scaling data pipelines.
- Proficiency in Python and experience with ML frameworks like Pytorch.
- Experience with tools and frameworks for ML Ops, such as Prefect, Kubeflow, MLflow, or similar.
- Solid understanding of software development principles and best practices in CI/CD for machine learning workflows.
- A generalist mindset, excellent problem-solving skills and the ability to collaborate effectively with cross-functional teams.
- Must be eligible to work in the US.