About the role
We are looking for a highly motivated and detail-oriented Principal Data Scientist to drive and shape the data science, advanced analytics, and visualisation capabilities at Dyson.
As a key member of the CITO team, you’ll work closely with internal teams to evaluate complex data, analyse data from multiple angles, train and deploy machine learning models, create impactful visualizations, and deliver findings that directly impact the business. Moreover, you’ll have the chance to impact how Dyson best executes data science globally, by supporting local data science teams with technical excellence guidance and acceleration.
As the Principal Data Scientist, you will play a key role in the delivery of high profile, high impact use cases through our squads. Moreover, you’ll have the chance to impact how Dyson best executes data science globally, by supporting local data science teams with technical excellence guidance and acceleration.
- You will be a technical pioneer, not only leading and delivering diverse, impactful data science projects across the organisation, but also offering consultancy to established data science teams, being a sparring partner to them, and supplementing their use case approaches with cutting edge techniques.
- You will architect, build and deploy predictive models using the myriad data sources available at Dyson, and enriching these with the creative sourcing of external data.
- You will apply logical thinking and statistical learning techniques to obtain robust results the business can rely on for critical decisions.
- You will lead the response to complex business questions beyond what business intelligence teams are capable of today.
- You will show up with ideas in your liaisons with the business, frequently guiding them and contributing to their advanced analytics roadmap.
- You will work alongside Innovation Architects, bringing your expert knowledge of machine learning and AI to discussions around the future technical landscape for Dyson.
- You will work with experts in CITO to get advice and support on accessing data and productionising your models.
- You will bring your experience of productionising models to Dyson, influencing and growing our MLOps capability, and ensuring that our models deliver the value they promise.
- Strong and demonstrable real-world experience of delivering tangible business value through data science models.
- Excellent business acumen - you are someone who can quickly absorb a new domain, and recognise what problems are most valuable for our stakeholders.
- Recognises the value of well commented, elegant, and test-driven code and demonstrate this in your work.
- A scientific and rigorous approach to solving practical problems using logical thinking.
- Strong core knowledge of applied machine learning supplemented with a robust statistical background.
- Hands-on experience with leveraging data from a wide selection of data sources from different technologies e.g., SQL, BigQuery etc.
- Finds the possibility of working closely in a squad with data architects, cloud enablers, data engineers and devops practitioners exciting.
- Keen understanding of data models and ETL processes. Using primarily Python and supplementing this with Tableau or Looker where necessary, you are able to analyse, model, and visualise data effectively.
- Ideally you will have at least a Master of Science qualification in a relevant field (e.g., Statistics, Mathematics), or similar hands-on experience.
- Fluent in Git and familiar with CI/CD practices to develop and deliver software optimally.
- Worked successfully using Agile methodologies or Kanban.
- Has a passion to understand Dyson, to frame its problems, and to deliver tactical solutions in short timeframes when required.
- Thrives on exercising your communication skills, with the ability to explain complex analytical concepts to business audiences of varying data literacy levels.
Essential Key Qualifications
- A degree in Statistics/Data Science/ML/Business Analytics or a science/engineering degree with a strong statistical element.
- PhDs are valued, but creativity and excellence to an even greater degree.
- Strong understanding of statistical modelling. Working experience of using advanced machine learning techniques to solve business challenges.
- Strong Python skills with a focus on statistical and ML packages, e.g., Scikit-Learn, TensorFlow, Keras, PyTorch, XGBoost, LightGBM, NumPy, SciPy.
- Working experience with cloud-based platforms. Comfortable querying modern cloud databases (e.g., BigQuery, Snowflake, Redshift).
- Successful use of software engineering best practices, including version control (Git, Mercurial), unit testing and working with Agile delivery principles.
- Proven track-record of using a rigorous, scientific approach to model building, testing and validation.
- Experience in data cleansing and blending (internally and externally) to drive richer insights.
- Knowledge of how to build enterprise data science products in cloud environments.
- Docker and Kubernetes experience.
- 3 years of programming experience in machine learning framework, TensorFlow, PyTorch or Keras.