Consultant, AI for Science – Knowledge Asset Design Product Development Division

PATH; Greater London; Remote; GBP 70,000 - 110,000; Full time

Job summary

PATH is a global nonprofit dedicated to health equity, leading an AI initiative to accelerate biomedical discovery and strengthen regulatory systems. This remote role oversees the expert process for a knowledge asset design within PATH's AI Initiative, under the Deputy Director, to advance vaccine science through evidence synthesis and informatics.

You will convene academic experts, draft background materials, run workshops, and produce an options paper for Gates Foundation funding with a clear

Qualifications

• PhD or equivalent in biomedical informatics or related field.
• Hands-on experience designing or building structured biomedical knowledge resources (knowledge graphs, databases).
• Familiarity with large language models and agentic systems.
• Experience leading expert consultations or technical working groups.
• Background in vaccine immunology or infectious disease preferred.

Responsibilities

• Review phase-one records and translate failure modes into knowledge asset requirements.
• Draft a conceptual outline for a minimum viable knowledge asset.
• Prepare background material for technical advisers and disease experts.
• Design and co-facilitate an expert workshop for automated extraction and scoring approaches.
• Summarise expert opinion, noting consensus, differences, and rationale.
• Draft an options paper detailing build priorities, impact, cost, and evaluation.
• Contribute to publications and presentations arising from the work.

Skills

Biomedical informatics

Knowledge graphs

Ontology-aligned databases

RAG/retrieval-augmented generation

Large language models

Expert consultations facilitation

Education

PhD or equivalent in biomedical informatics or related field

Tools

RAG systems for literature

Ontology curation tools

Job description

Here PATH is a global nonprofit dedicated to achieving health equity. With more than 40 years of experience forging multisector partnerships and with expertise in science, economics, technology, advocacy, and dozens of other specialties, PATH develops and scales up innovative solutions to the world's most pressing health challenges.

This is a global remote opportunity.

PATH's AI Initiative is pioneering the use of generative AI and other advanced digital technologies to accelerate biomedical discovery, strengthen regulatory systems, and enhance the safety and scalability of digital healthcare.

About the role:

PATH leads a three-year programme, funded by the Gates Foundation, to test whether an AI co-scientist can help identify potential correlates of protection for vaccine development and shorten the path to licensing new and improved vaccines. The programme began with rotavirus and maternal RSV vaccination and will move to other diseases in its final phase. In phase one, PATH used off-the-shelf co-scientist tools built by frontier AI labs to generate and rank candidate biomarkers for rotavirus and maternal RSV vaccination, with PATH scientists and external experts reviewing each cycle. Those reviews showed that the tools’ main limitations were shaped by the scientific knowledge they could access and how they weighed evidence on immunology, assays and trials. Phase two will focus on designing a small, open knowledge asset that measurably improves how an agentic tool generates and ranks vaccine science hypotheses. Working under the supervision of the Deputy Director (AI4Science), the consultant will lead the expert process that produces this design. You'll help convene a technical advisory group of academic experts in evidence synthesis, biomedical data structuring and informatics. You'll work alongside PATH's own AI specialists and bring the group together with vaccine scientists who work on correlates. You'll draft the material the experts respond to, run the conversations, and take part in them as one of the technical voices. By the end of 2026 you'll turn their views into options for a decision with the Gates Foundation on which parts of the asset to build first.

What you'll be doing:

• Work with the PATH team to critically review the phase-one record and translate observed failure modes into an initial view of what the knowledge asset must contain.
• Draft a conceptual outline for a minimum viable knowledge asset.
• Brief and engage technical advisers and disease experts, individually and as a group, preparing the background material needed for each conversation.
• Design and co-facilitate an expert workshop, either as a single convening or a series of sessions, to agree methods for automated extraction from biomedical sources and approaches for scoring and prioritising knowledge, building on an evidence-quality scoring framework PATH has developed in related work.
• Reconcile and summarise expert opinion, noting where consensus was reached, where views differed, and the evidence or reasoning behind each position.
• Draft an options paper that sets out what to build first, the expected impact, cost and complexity, and how each element would be evaluated, with enough detail to scope the build that follows.
• Contribute to publications and presentations arising from the work, where the results warrant it.

By December 2026 you'll have produced:

• Briefing material for the advisers and experts
• A conceptual outline of the minimum viable knowledge asset
• A workshop design and a written synthesis of its conclusions
• An options paper to support the funding decision, which will form the basis of the programme's knowledge asset roadmap.

Required skills and experience:

• PhD, or equivalent research experience, in biomedical informatics, computational biology, bioinformatics or a related field.
• Hands-on experience designing or building structured biomedical knowledge resources, such as knowledge graphs, ontology-aligned databases, curated evidence tables or retrieval-augmented generation systems for scientific literature.
• Working familiarity with large language models and agentic systems, including how retrieval and tool access, for example through the Model Context Protocol, shape what a model produces.
• Experience running expert consultations, technical working groups or similar processes, and a record of synthesising their conclusions.
• Ideally, grounding in vaccine immunology or infectious disease sufficient to follow and contribute to technical discussions.

We know great candidates won't always meet every listed qualification. Research shows that some groups, on average, are more likely to self-select out if they feel they don't meet all requirements.

We're searching worldwide for experienced leaders and passionate emerging professionals in many disciplines - from science and engineering to accounting to information technology - who share our belief that better health moves humanity forward.