Location
London

Hours
Not specified

Salary
Negotiable

About the Role
Knowledge is becoming a source of competitive advantage. Organisations that have organised, current and trusted knowledge, usable by both people and AI, will make better decisions, move faster, and serve customers better than those that do not. We want our knowledge to be a quiet advantage for ClearScore Group, not a tax.

This role exists to do two things at once. The first is hands-on. We have an ever-growing set of working documents, Notion pages, dashboards, and content across a cloud drive, SharePoint and Slack. Some of it should no longer exist. Some of it is foundational and is at risk of being lost. Someone needs to roll their sleeves up: write, rewrite, restructure, decommission, and own the canonical index of what we actually know. This is part librarian, part documentarian, part data steward.

The second is strategic. We need to stop treating our corpus of working documents as if it were authoritative knowledge, build a deliberate corpus that is, maintain it actively as facts change, fill the gaps in critical domains like products, regulation and markets, and make sure every piece of significant work leaves behind reusable learning. This is the knowledge architect side of the role.

The person in this seat owns the standards, owns the canonical index, does the work where it has to be done by them, and gets the rest done through a federated network of domain owners and maintainers across the Group.

This is a mid-level individual contributor role with no direct reports. It sits centrally and reports to the VP of Operations who reports directly to the CEO. The reporting line is deliberately function-agnostic: not inside Data, not inside Architecture, and not inside AI Engineering. This is so the role can set standards across all of those functions without being captured by any of them.

Key Responsibilities:
1. Produce and roll out a written Knowledge Strategy endorsed by the executive, explicitly addressing how we do knowledge management in an AI-enabled world.
2. Separate knowledge from working documents by defining what counts as knowledge for the Group (authoritative, owned, current, trustworthy content) versus working documentation such as drafts, scratchpads, meeting notes and in-flight thinking. Set a higher bar for publishing to the knowledge corpus and use LLM-based tools to help enforce it.
3. Actively maintain the corpus by setting review cycles, ownership and freshness expectations, running the rhythm that keeps them honest, surfacing and resolving stale, duplicate or contradictory content using LLM-based tools, treating decommissioning as a first-class activity with a credible first pass at the existing estate inside six months, maintaining clear lineage, and working with Legal, Risk and Compliance on access, retention and regulator-facing obligations.
4. Fill gaps in critical domains by mapping domains across the Group, identifying where authoritative knowledge is thin or missing, and working down a prioritised backlog through a federated maintainer model with named experts and maintainers.
5. Make compounding learning a habit by establishing a simple, mandatory pattern for every significant piece of work to leave behind structured learning, making it the default output of quarterly business reviews, retrospectives, incident washups, experiments and major decisions, and curating the resulting record for retrievability by people and AI tools.
6. Equip our knowledge for AI consumption by understanding how modern AI systems consume organisational knowledge (retrieval, structure, metadata, permissions), specifying what good looks like from the knowledge side, and working with AI Lead, Chief Architect, Chief Data Officer and Engineering on indexing, retrieval and serving knowledge into internal AI tools.
7. Roll up your sleeves to author and maintain the Group's canonical knowledge index: the single place any employee or internal AI tool goes to find the authoritative source on a topic.

Experience
- Proven track record of materially improving how knowledge is organised, owned and maintained in a mid-to-large organisation through strategy, standards and hands-on work.
- Comfortable with data governance concepts as applied to knowledge: ownership, custodianship, classification, access, lineage, retention.
- Awareness of obligations in regulated markets.
- Practical understanding of how modern AI tools consume organisational knowledge, including retrieval, structure, metadata and permissions.
- Fluent across tooling landscapes such as wikis, documentation systems, knowledge bases, document stores, and search.
- Experience influencing without formal authority and managing federated networks of maintainers.

About You
- Brings order to complexity and is comfortable operating across functions without formal authority.
- Bias to action; happy to write, fix, rename, restructure and archive as needed.
- Strong writer and structurer able to turn messy source material into clear, accurate, well-structured content.
- Able to say no, hold standards, and stay constructive.

Qualifications
- Technical depth can be grown into; judgement and bias to action are essential.
- Desirable: Experience in financial services, fintech or regulated industries.
- Desirable: Background in library or information science, technical writing, knowledge engineering, records management or data governance.
- Desirable: Experience operating a federated model with appointed experts and maintainers.
- Desirable: Exposure to API-led businesses and organisations transitioning to AI-assisted retrieval on the knowledge side.