Police AI - Head of Data Science and Evaluation

This post is open to applicants who can meet the College’s nationality, UK residency and vetting requirements.​ 

​​Remote working (anywhere in the UK). Please note, some travel may be required.​ 

PoliceAI is a nationally significant technology and transformation programme in UK policing. Backed by central government investment and positioned at the heart of the police reform agenda, PoliceAI is the new National Centre for AI in Policing, helping to shape how policing uses data, technology and artificial intelligence to tackle crime, protect the public and modernise frontline operations. 

We bring together operational policing, technical expertise, product delivery, assurance and the responsible use of AI to develop and scale capabilities that improve public safety outcomes. Our work helps forces respond more effectively to demand, strengthen investigations, identify risk and harm earlier, and free up frontline time for the activities that matter most to communities. 

This is an opportunity to contribute to a programme with national reach and meaningful public service impact, helping to define how AI is developed, governed and adopted across one of the UK’s most important public services. The work is varied and complex, moves at pace, and has potential to deliver better tools for policing, better outcomes for the public, and stronger foundations for the responsible use of AI. 

Police AI is hosted within the College of Policing.    

Our mission is: Leadership, Standards, Performance. This guides our work with individuals, forces, and policing partners towards our vision. We exist to support police officers, police staff and volunteers to deliver the best service to the public. Only through high quality leadership, consistent standards and continual performance improvement can everyone in policing reach their full potential. Our mission guides our work with individual, forces and partners towards our vision. 

Additional Information:   

We recruit by merit on the basis of fair and open competition, as outlined in the Civil Service Commission's recruitment principles. 

Due to our status as Public Servants, we are not eligible to participate in the Civil Service transfer process. Therefore, while successful candidates can transfer in on their existing terms and conditions, any subsequent transfer out of the College to the Civil Service cannot be guaranteed on existing terms and conditions, and new starter/modernised terms may apply.  

The College embraces diversity and promotes equal opportunities. As such, we comply with the Disability Confident Scheme (DCS) and apply the minimum selection criteria at sift stage of the recruitment process.   

We are looking for people who are motivated by public service, interested in responsible innovation, and comfortable working collaboratively across organisational, operational and technical boundaries. Joining PoliceAI means contributing to a supportive, multidisciplinary team that is helping to shape the future of policing and ensure AI is used in ways that are effective, responsible, operationally useful and worthy of public trust. 

As Head of Data Science & Evaluation, you will balance immediate delivery with the long-term task of building a high-performing data science and evaluation function. Leading the development and rigorous assessment of AI-enabled products that improve public safety, productivity and decision-making, you will also shape the team, methods, standards and evidence base that allow Police AI to innovate at pace and earn trust over the years ahead. 

You will: 

• Build, lead and grow a high-performing multidisciplinary team of data scientists and testing and evaluation specialists. 
• Establish repeatable and auditable testing, evaluation, verification and validation (TEVV) frameworks, spanning dataset provenance and quality, robustness, bias, human–computer interaction, wider system effects and lifecycle monitoring. 
• Lead the internal development and assessment of AI solutions, translating operational policing problems into testable hypotheses, meaningful measures and practical products. 
• Define and implement the data foundations required for credible evaluation, including ground truth data requirements, labelling processes, provenance and versioning, synthetic data generation, test-set construction and data quality controls. 

We are looking for someone with a strong record of leading data science teams and delivering AI solutions, who is comfortable developing people whilst embedding the standards and practices needed to evaluate AI rigorously and scale its use responsibly. 
 
You will combine deep technical credibility with pragmatic leadership, working under ambiguity, navigating difficult methodological choices and a varied range of stakeholders to deliver material improvements across policing. You will work closely with engineering, delivery, data governance and Responsible AI colleagues, as well as operational users and partners, to turn complex needs into useful solutions and defensible evidence. 
 
You will be expected to take well-founded decisions on areas such as model and data strategy, evaluation design, centralised and distributed testing, open-source and commercial solutions, and continuous monitoring, as well as providing input into development priorities and direction. Communication of performance, limitations, trade-offs and risk must be clear to technical and non-technical audiences. 
 
For the right person, this role offers a compelling challenge: technically difficult and varied data science, ownership of a nationally influential evaluation function, direct public impact and the chance to raise the standard for how AI is developed, tested and adopted across policing. 

Some of your responsibilities will include: 

• ​​Building and developing a high-performing data science and evaluation team; coaching Data Scientists and evaluation specialists, setting clear objectives, and embedding a culture of innovation, methodological rigour and technical excellence. ​ 

• ​​Leading technical delivery of AI-enabled solutions and their evaluation across the full lifecycle, from problem definition, model and data strategy and experimentation through to deployment, monitoring and continuous assessment.  ​ 

• ​​Defining consistent, repeatable and auditable TEVV approaches, working with Engineering to build reusable testing pipelines and sound foundations for dataset provenance, versioning, sampling, labelling and data quality.​  

To be successful in this role you’ll need: 

• ​​Proven leadership of data science or AI teams, including hiring, line management, coaching and development, with experience of establishing, growing and scaling technical capability and leading teams through delivery in line with organisational goals, user needs, priorities and timescales.  ​ 

• ​​Significant hands-on experience of developing and deploying production-grade AI/ML and generative AI solutions, with strong technical judgement, coding practice and proficiency in a range of data science domains.  ​ 

• ​​Expertise in proportionate AI testing and evaluation, including validation and verification, robustness and adversarial testing, bias and fairness, interpretability, human factors and generative AI challenges such as hallucination, prompt sensitivity, guardrail testing and RAG evaluation.  ​ 

Please see the attached job description which contains full details of the role. 

We'll assess you against these behaviours during the selection process:

• Managing a Quality Service
• Making Effective Decisions
• Delivering at Pace
• Seeing the Big Picture