We build the data foundations that make AI useful and safe inside regulated financial firms. The value of AI is capped by the data its agents can reach: if an agent cannot find, interpret, trace or be correctly permissioned against data, the capability is useless, or worse, unsafe. Your job is to close that gap.
Our client is a leading global investment management company headquartered in London. It manages over $228 billion in assets and serves institutional investors, pension funds, wealth managers, and other sophisticated clients worldwide. The firm specializes in quantitative investing, alternative investments, systematic trading strategies, and technology-driven asset management. Data science, machine learning, and AI are core components of its investment and research processes.
- Requirements:8+ years in data architecture or data platform roles, including 3+ years in metadata management, data governance or data cataloguing across a large and diverse data estate.
- Architectural experience with enterprise data catalogues (e.g. DataHub, OpenMetadata, Collibra or similar); understanding of the concepts matters more than any specific product.
- Solid knowledge of metadata and lineage standards (e.g. OpenLineage, Open Data Contract Standard or similar) and experience integrating with proprietary in-house metadata and event models.
- Proven design of federated (hub-and-spoke) metadata architectures: a central layer for identity, hierarchy and links, domain-level catalogues with rich local detail, and a clear contract between the two.
- Practical experience with data governance operating models: ownership and stewardship roles, sensitivity classification, metadata quality measurement.
- Semantic layer design: business glossaries and concept registries, resolution of term conflicts between departments, entity resolution, knowledge graph modelling; design-level knowledge of graph databases.
- Strong consulting and client-facing skills: leading working sessions, defending design decisions in written reviews, negotiating boundaries between teams with overlapping catalogue initiatives.
- Able to produce client-ready architecture documents and written review responses without editorial support.
Will be a plus:
- Knowledge of financial-industry ontologies (e.g. FIBO or similar) and a realistic view of their practical limitations.
- Familiarity with semantic search over metadata based on embeddings.
- Domain experience in asset management, market data or fund reporting.
- Experience in on-premise or regulated environments: data residency, auditability, licence-scoped data entitlements.
- Experience designing metadata and discovery layers consumed by AI agents.
- Pre-sales or discovery and solutioning experience; experience joining an engagement already in progress.
Responsibilities:
- Own the federated metadata architecture: the central discovery layer, the domain catalogues for market data and curated reporting, and the federation contract that binds them.
- Design domain metadata models in working sessions with the client teams that own the data, extending the client's existing catalogue model instead of replacing it.
- Define the approach to machine-derived metadata: what is harvested automatically, what is drafted by LLMs and approved by human stewards, what remains manual, and how its quality is scored.
- Design the semantic layer: shared business-term definitions with per-department mappings, entity resolution across sources, and the knowledge graph that supports guided discovery.
- Act as design authority for the engineering pod: review integration designs and keep parallel implementations aligned to one architecture.
- Represent the design in client governance: reviews, written responses to senior stakeholders, coordination with the client's own initiatives and with the parallel entitlement and security workstream.
- Shape phased delivery plans, effort estimates and data-readiness prerequisites for the implementation phase.
