We are looking for a data scientist to help us develop the state-of-the-art algorithms we use for robot control – leveraging our large fleet of robots to gather data and validate improvements. We are open to candidates with a relatively broad range of experience levels, including recent graduates, perhaps with experience gained in the context of a PhD. We prize intelligence and good intuition about the practical challenges inherent in performing experimental science using large datasets.
Duties & Responsibilities
- Sampling from the huge amount of image data obtained by our fleet of soft fruit picking robots to create training/validation/test datasets
- Creating the data pipelines required to allow us to manage and use large image datasets (storing, indexing, accessing, etc.)
- Helping to define ML training process and experimental design
- Developing principled scientific methods for validating algorithmic improvements in the field – in the context of a large fleet of soft fruit picking robots
- Ownership of image annotation processes and tools
- Engaging with other technical to develop processes for logging, data capture, etc.
Desirable Characteristics
- Understands and can apply machine learning methodology
- Good intuition about how to manage and use massive datasets effectively in scientific applications
- A high quality degree (a PhD in a scientific discipline is appealing but not required)
- Experience of using Python and/or C#
- A willingness to learn new skills