Data Science Lead

Title: Data Science Lead Location: Abingdon, UK

Data Science Lead | Semiconductor Metrology
Abingdon, UK

Reward and Benefits

• £70,000 - £85,000 (depending on experience)
• Annual bonus, 3x salary death in service benefit and private health care BUPA, EMI Share scheme.

If you are a Data Science Lead who enjoys turning measurement data into proof of what is possible, why not join us in Abingdon, Oxfordshire, and build and lead the Data Science team behind advanced semiconductor metrology systems.

We are looking for an ambitious and adaptable Data Science Lead to set up a new Data Science team in our Technology department, reporting to the Technology Director. You will lead a small team, initially a Data Scientist and a Modelling Scientist, and stay hands-on as a working lead data scientist. You will explore and prove new analysis methods, from synthetic data and information-limit studies to modelling, machine learning and AI, to improve the performance of our Metron3D systems and draw more insight from our data.

You will work hand in hand with the Data Analysis team in Engineering: you prove what is possible, they make it robust and shippable. You will also collaborate with R&D, System Design Engineering and Applications Development. We are an AI-first team, and you will set the standard for using AI tools well and safely. If your passion is speed, evidence and finding out what the data can really tell you, we’d love to hear from you.

What you’ll be doing

• Lead, develop and mentor the Data Science team while staying hands-on, setting priorities and keeping a fast, evidence-based working rhythm
• Own and deliver the Data Science roadmap: synthetic data, modelling, information-limit studies, algorithm development and data analysis tools
• Establish the information limit, meaning what is physically extractable given instrument noise, limitations and tip geometry, and use it as the benchmark for every new method
• Explore, test and benchmark new algorithms (for example PCA, percentile methods and machine learning) on real and synthetic data, covering achievable performance, measurement uncertainty and gauge capability
• Define the handover to Data Analysis in Engineering (prototype code, test data, acceptance criteria and expected performance), and manage the demo and performance data libraries with Applications Development
• Use AI tools every day, set the team’s standards for using them well and safely, and track the state of the art in third-party and open-source methods

Education and experience

• MSc or PhD in Physics, Data Science, Applied Mathematics, Engineering or a similar subject, or [8] years’ relevant experience in data science or quantitative research, including some team leadership

Skills and experience we’re looking for:

• Experience leading or mentoring a team of data scientists, scientists or engineers
• Hands-on data science in Python (NumPy, SciPy, pandas, scikit-learn or similar), turning exploratory work into quantified results
• Heavy, day-to-day use of AI tools in your own technical work, such as coding assistants, LLMs and agentic workflows, with a critical eye for checking their outputs
• Image processing, signal processing or statistical analysis of measurement data from scientific instruments
• Statistical methods for measurement systems: uncertainty, repeatability, gauge R&R and drift
• Machine learning and/or physics-based modelling, with the judgement to pick the right tool for the problem
• Clear communication of results to technical and non-technical audiences, with evidence-based recommendations
• Comfortable setting a roadmap and priorities, and knowing when a result is good enough to hand over

We’d love to see:

• Atomic Force Microscopy or other nano-characterisation (SEM, FIB, profilometry), or semiconductor metrology and process control
• Synthetic data generation or forward modelling of instruments
• Building AI-assisted workflows or tooling for a team (agents, automated analysis pipelines, shared prompt libraries), and handing prototypes to software teams to productionise