AI Engineer in Reading

We’re hiring an AI Engineer to join a fast-growing UK-based AI product company building intelligent auto-assessment systems for education and workforce certification. The role is fully remote within the UK, with a preference for candidates based in or near London. You’ll work on production-grade AI that automates the evaluation of knowledge and skills using real customer data such as marking rubrics, example answers, SME feedback, and historical scoring patterns. This is a hands-on engineering role with full ownership of the AI model lifecycle, from data preparation and experimentation through to deployment, observability, and continuous improvement, in a small, pragmatic start-up environment.

Technical requirements and experience:

• 5–6 years of AI or ML engineering experience, or a relevant PhD with 2+ years industry experience
• Strong Python skills and experience building APIs
• Hands-on experience with LLMs, agentic workflows, and frameworks such as LangChain, LangGraph, DSPy, or similar
• Proven experience owning the full model lifecycle in production
• Experience designing evaluation frameworks and statistically comparing model performance (e.g. t-tests)
• Experiment tracking and observability tools such as MLflow, W&B, DVC, Langfuse, or similar
• Cloud experience, ideally AWS (Azure also considered)
• Experience with Git and collaborative development workflows
• Understanding of data privacy, fairness, and scalable AI system design

This role is ideal for an experienced AI or ML engineer who enjoys autonomy and impact. You’ll design evaluation frameworks aligned to real assessment criteria, experiment with and benchmark LLMs, optimise cost and latency, and build multimodal workflows that analyse text, images, and video. You’ll work closely with engineers and assessment experts, help shape the AI architecture, and build systems that generate accurate scores alongside clear, actionable feedback for learners and educators. There’s a strong focus on quality, fairness, auditability, and statistical rigour in model evaluation.

You should want to work here if you’re motivated by meaningful AI problems, real-world impact, and technical ownership. You’ll have the freedom to make decisions, influence product direction, and see your work used in high-stakes environments. The company offers a competitive salary, equity or stock options, pension contributions with employer match, 25 days annual leave, and a company MacBook, all within a supportive and ambitious start-up culture focused on long-term impact rather than hype.