Position Overview

The AI Scientist (Advanced Specialist) is a recognised subject-matter expert who delivers high-impact machine learning solutions and helps business units across Pearson make better use of existing AI capabilities. The role sits at the intersection of hands-on modelling and practical problem solving with partners: this person is as comfortable working through a use case with a product team as they are reviewing a model implementation or proposing a new technical approach.

A significant part of this role is about connecting the dots. Pearson has a growing set of AI components and capabilities across different products and platforms. The Advanced Specialist actively works to understand what exists, identify where it could be applied again, and make the case to teams that could benefit. This is not a separate advisory function — it is embedded in the way the role operates day to day, alongside its core technical responsibilities.

Key Responsibilities

• Deliver robust, scalable ML solutions for AI capabilities in educational products, taking significant implementation decisions independently.
• Work closely with Business Unit partners to understand their AI-related problems, help frame them clearly, and develop solutions that fit their actual needs.
• Identify opportunities to reuse existing AI components and capabilities across teams. This includes use case discovery, technical solutioning, and making a credible case to partner teams for why reuse is the right path rather than building from scratch.
• Review and validate the outputs of more junior scientists, providing expert-level technical feedback and catching issues before they reach production.
• Shape the direction of projects within the team, bringing experience and a point of view to decisions about approach, tooling, and scope.
• Propose new techniques and methods where current approaches have clear limitations, and take responsibility for evaluating and introducing them.
• Mentor peers and support the growth of colleagues at earlier career stages, through code reviews, pairing, and direct feedback.
• Maintain thorough documentation and uphold the technical health of deliverables over time, not just at point of delivery.

Expected Deliverables

• Production-quality models and algorithms that improve measurable outcomes for the products and business units they serve.
• Documented reuse cases: instances where an existing AI component was identified, evaluated, and successfully applied in a new context.
• Expert peer reviews that add genuine value to the team's work and improve the quality of what gets shipped.
• Clear technical documentation that makes solutions maintainable and understandable to others.
• Positive, demonstrable impact on the growth of colleagues through mentoring and knowledge sharing.

Required Qualifications

• Strong applied experience in machine learning and deep learning, with a track record of delivering solutions that have gone into production and affected real outcomes.
• Effective proficiency in Python and core ML frameworks (e.g. PyTorch, TensorFlow, Hugging Face, scikit-learn), including model evaluation, experimentation, and iteration at scale.
• Experience working directly with non-technical or semi-technical stakeholders to understand problems, frame requirements, and align on the right technical approach.
• Ability to assess an existing capability or component critically and make a reasoned case for whether it fits a new context.
• Comfortable shaping project direction and making significant technical decisions with appropriate autonomy.
• Experience reviewing others' work at a high level of rigour, and providing feedback that improves both the output and the person who produced it.

Preferred Qualifications / Nice to Have

• Familiarity with educational technology, language learning, assessment, or speech processing domains.
• Experience with generative AI, including LLMs, fine-tuning, RAG, or evaluation of LLM-based systems.
• Background working in or alongside a platform, product, or enablement team where reuse and shared components were an explicit consideration.
• Exposure to responsible AI principles in practice: fairness, bias, explainability, and how these apply in education contexts.
• Experience working across multiple teams or business units simultaneously rather than within a single product.