Ordnance Survey (OS) provides national mapping services for Great Britain and is world-leading provider of geospatial solutions.
OS location data and expertise has helped governments make smarter decisions, businesses gain valuable data insight, and everyone experience the world outside for over 225 years.
Our busy Propositions and Innovation team is looking for people to help inspire the organisation and our customers, to go ‘somewhere new’, by building a portfolio of next generation services.
Join us as we transform OS, deliver commercial growth and empower our customers to meet their goals. Most of all, you’ll play a critical role in helping OS to realise its vision: to be recognised as world leaders in geospatial services; creating location insight for positive impact.
About the role
Ordnance Survey has a proud history of investing in research and innovation that develops novel approaches to extracting information from imagery. We are looking for someone with expertise in this area to help support a technical program in the deep learning image analysis space. You will develop upon recent advances in computer vision and deep learning to radically change the way we operate, enabling new products and services for both our domestic and international customers.
Ordnance Survey has been working in the field of deep learning for many years and has created new capability that extracts insights from imagery through a combination of inference and discovery. You will be responsible for transitioning knowledge from external research into OS and developing, testing, prototyping, and evaluating methods and algorithms. You will also work closely with engineering teams to help robustly scale deep learning and ensemble models in production.
You will be a part of the Innovation team within Ordnance Survey’s Proposition and Innovation (P&I) Group - a new function within OS which consists of over 100 professionals, delivering a world leading research, innovation, propositions, data, and product management capability. P&I will be utilising disruptive technology of which Earth Observation (EO) will be a key enabler, to enter new markets, create new breakthrough products, whilst delivering medium to long term commercial objectives.
With a focus on deep learning, design and implement scalable geospatial data analysis solutions to meet specific data requirements and discover new insights.
Create, run, and document experiments and communicate results to key stakeholders.
Develop research code in Python and potentially other languages (C/C++, Scala).
Prepare data for further analysis using data cleansing, data fusion and feature engineering techniques.
Identify and develop new opportunities to exploit and extract value from Earth Observation imagery and other geospatial data to solve real-world problems for domestic and international customers.
Provide thought leadership and drive best practice.
Keep current with new technologies and research developments.
You will ideally hold a degree qualification in computer science, data science, mathematics, physics, or a similar analytical subject. Alternatively, you will have significant experience working in a relevant field within a research or production environment.
What we’re looking for
In-depth familiarity with machine learning approaches to image analysis, using both supervised and unsupervised techniques.
Experience developing deep learning models for image analysis with a focus on transfer learning, efficient semantic/panoptic segmentation, and ensemble approaches.
Detailed understanding of the deep learning model lifecycle, MLOps and metrics associated with model performance and interpretability.
Idiomatic use of Python and knowledge of any (or all) of the following frameworks and platforms: Tensorflow, Keras, PyTorch, Spark, AzureML, DataBricks.
An ability to manipulate, combine and present raster and vector data, using a range of tools (either off-the-shelf or custom built).
Knowledge of DevOps best practices relating to continuous integration/deployment and version control.
Significant experience working with cloud platforms such as Azure, AWS or GCP.
Ability to work and collaborate with an interdisciplinary team of developers, research scientists, data scientists, data engineers and subject matter experts from OS and partner organisations.
Excellent communication and presentation skills across a wide range of audiences.
Ideally you will have knowledge and experience of:
Working with geospatial data and GIS software (Esri, FME, eCognition).
Experience of combining Earth Observation and remote sensing data with geospatial data to solve complex problems.
Familiarity with data visualization tools relevant for image analysis and geospatial data.
We want you to love what you do. That’s why our benefits package rewards a job well done. We’ll give you:
Salary – £36,414 - £40,356
Performance related bonus
A competitive pension scheme
37 hour working week (with flexible working options)
25 days annual leave - (30 days after five years) bank holidays and an extra 3 over Christmas
Plus, a suite of excellent additional benefits
Location: This role will be based at our state-of-the-art Headquarters in Southampton. However, all our people, where they can are currently working from home.
Closing date: Friday April 23rd 2021
We believe diversity and inclusion is about working together – in a supportive and respectful environment to reach our full potential. We think combining different backgrounds, experiences and perspectives will help us reach our vision and be trusted and admired across the globe for setting the standards and leading the way. We are looking for talented and experienced individuals from a range of backgrounds and welcome applications from any race, age, gender, background or religion.
We’re individually talented and collectively powerful and we give you the space to take your career in whichever direction you want.
We’re always happy to hear from you, so don’t forget to follow us on Twitter, LinkedIn or find us on Facebook.