AI Evaluation Engineer PSAI Tech - Central Government

Role: AI Evaluation Engineer PSAI Tech

Contract Length: 4 months

Location: Hybrid/London, Bristol or Manchester - expected to attend the two days per week

IR35: Inside

Pay Rate to Intermediary: £700-£750 per day

Security Clearance: BPSS and SC eligible requested

Spinwell is recruiting for an AI Evaluation Engineer PSAI Tech for an excellent opportunity within the public sector.

RESPONSIBILITIES OF THE AI EVALUATION ENGINEER PSAI TECH

• act as an AI Evaluation Engineer embedded with business and product teams to support the development and deployment of risk management toolkit.
• translate business requirements, risk-tiering frameworks and organisational risk tolerances into measurable evaluation criteria.
• design, implement and maintain AI evaluation methodologies, test suites and quality assurance processes with the downstream application teams.
• define baseline performance metrics and risk thresholds for AI products and services.
• assess model quality, reliability, robustness, safety and alignment against agreed standards for the downstream usage.
• help teams develop automated and manual evaluation pipelines to measure product performance across a range of scenarios.
• identify, investigate and communicate evaluation findings, risks and quality issues to technical and non-technical stakeholders.
• work closely with data scientists, engineers, product managers and policy teams to improve AI system performance.
• produce evaluation reports, technical documentation and recommendations to support governance and release decisions.
• contribute to the development of AI assurance, testing and quality engineering best practices across PSAI Tech.
• support audit, compliance and assurance activities relating to AI products and services.

SKILLS/EXPERIENCE OF THE AI EVALUATION ENGINEER PSAI TECH

• proven experience in quality engineering, testing or evaluation of data-driven and AI-powered products.
• strong understanding of machine learning systems, large language models (LLMs) and AI evaluation methodologies for production.
• experience defining performance metrics, KPIs, benchmarks and acceptance criteria for AI products.
• ability to translate business risks and policy requirements into measurable technical evaluation frameworks.
• experience developing automated testing and evaluation pipelines using Python or similar technologies.
• knowledge of AI safety, model robustness, bias, fairness and responsible AI principles.
• strong analytical and problem-solving skills with the ability to identify trends, risks and performance issues.
• experience working with cross-functional teams in agile product delivery environments.
• ability to communicate complex technical findings clearly to both technical and non-technical audiences.
• experience producing technical reports, dashboards and recommendations for senior stakeholders.
• familiarity with cloud-based AI platforms and modern software development practices.
• experience working in government, public sector or regulated environments is desirable.

We welcome all applications regardless of background, in line with our commitment to diversity, equality and inclusion.

Applying to this or any other vacancy advertised by Spinwell Ltd constitutes an agreement for Spinwell Ltd to hold your details for 24 months for the purpose of assessing suitability for the advertised position and to make you aware of any other positions deemed suitable of which You will make you aware by means of either email, text or phone. In line with GDPR regulations you are able to request your details be removed from the company data at any time by emailing us

#J-18808-Ljbffr