Artificial Intelligence and Machine Learning Data Scientist
Have a passion for AI and machine learning? Want to join a team of internationally renowned scientists dedicated to addressing some the world's greatest environmental challenges?
Internationally renowned marine research organization, Plymouth Marine Laboratory (PML) has an exciting opening for an Artificial Intelligence and Machine Learning Data Scientist.
PML is known for delivering impactful and award-winning science, from the poles to the tropics, and from lakes to the open ocean. With world-class research capability in global Earth Observation, ecosystem modelling and marine ecosystem functioning, its core research program specialises in areas including climate change, marine pollution and sustainability.
PML has one of the largest aquatic remote sensing groups in the world with 40 permanent staff, undertaking both fundamental research and operational regional and global EO data processing. The group hosts the NERC Earth Observation Data Acquisition and Analysis Service (NEODAAS; ) to provide EO data and services to the UK environmental communities.
NEODAAS was awarded a GBP1 million NERC "transformational" capital grant to purchase the Massive Graphics Processing Unit Cluster for Earth Observation (MAGEO) to facilitate application of Deep Learning to Earth Observation data. This represents the largest supercomputer of its kind dedicated to AI applications using earth observation. The cluster was installed in August 2020 and is built around 5 NVIDIA DGX-1 MaxQ nodes, providing a total of 204,800 CUDA cores with a dedicated 2 PB of storage, split into 1.5 PB NAS and 0.5 PB Lustre filesystem.
This position offers the opportunity to develop new research and applications for AI, utilising the MAGEO cluster. Working in collaboration with end-users and the rest of the NEODAAS and wider PML team, the post holder will address scientific questions through the application of appropriate Deep Learning algorithms to a range of areas supported by NERC (including terrestrial, atmospheric and marine). The post holder will also work towards increasing the utilisation of AI by scientists within PML and externally. For example, through contributing to documentation and training material on the use of Deep Learning on the MAGEO system which will be used by NEODAAS end-users and internally at PML.
Experience & Eligibility Requirements
- Masters or PhD based around applied Deep Learning or equivalent industry experience
- Computer vision/image segmentation and object detection experience with across multiple problem areas.
- An enthusiasm for working with others to solve problems using Deep Learning.
- Strong development skills in Python, using key Deep Learning libraries such as Tensorflow/PyTorch.
- A passion for producing well designed and documented code.
- Use of Linux systems for Data Science tasks
- Sufficient numerical ability to understand some fundamental theory behind deep learning algorithms. Strong statistical skills are particularly relevant.
The following skills would also be beneficial:
- Experience using Earth Observation data
- Experience with High-Performance Computing using multiple GPUs
- Use of container technologies such as Docker, Singularity and Kubernetes
- Familiarity with distributed learning approaches
PML operates a hybrid working policy whereby employees are able to split their working arrangements between PML's Plymouth offices and home-based working.
PML is committed to equality, diversity and inclusion, and our policy can be found on our website. We are proud to have achieved the Athena SWAN award as recognition of our achievements in gender equality. As part of this, we offer opportunities to discuss flexible working and whilst the selection process will be based on merit, we particularly welcome applications from female candidates, currently underrepresented.
PML also offers a variety of employee benefits.
*PML offers annual salary increments over a five year period, rising to GBP38486 (Scientist Grade)/GBP46410 (Senior Scientist Grade) maximum.