Assistant Vice President, AI Data Engineer (Semantic Layer) Greater Singapore Area

Temasek is a global investment company headquartered in Singapore, with a net portfolio value of S$518 billion (US$401b, €350b, £304b, RMB2.77t) as at 31 March 2026. Our Purpose “So Every Generation Prospers” guides us to make a difference for today’s and future generations. We seek to build a resilient and forward-looking portfolio that will deliver good sustainable returns over the long term. We have 13 offices in 9 countries around the world: Beijing, Hanoi, Mumbai, Shanghai, Shenzhen, and Singapore in Asia; and Brussels, London, Mexico City, New York, Paris, San Francisco, and Washington, DC outside Asia. For more information on Temasek, please visit For Temasek Review 2026, please visit Sustainability Report 2026, please visit

AI agents are only as good as the data they can reason over, and they can only reason over it reliably when they understand what that data means. Inconsistent metric definitions, ambiguous business terms and poorly governed data are among the most common reasons enterprise AI products fail to give accurate, trusted answers. The gap is usually data readiness and business context, not model capability. The Snowflake Semantic Layer Lead at Temasek is responsible for building the semantic foundation that makes Temasek's AI agents and analytics accurate, consistent and trustworthy across the complex, heterogeneous data environment of a global investment institution.

This role sits at the intersection of data engineering, data modelling and AI systems engineering. You will design and build the semantic layer on Snowflake that defines what Temasek's metrics, entities and relationships mean, along with the pipelines that feed it, so that AI agents, BI tools and analysts all draw on a single, governed source of truth. You will work across structured investment data (portfolio positions, valuations, financial statements, market data) and help connect it with unstructured sources (research reports, company filings, documents), so that agents can retrieve, reason over and act on Temasek's data with confidence.Responsibilities

You will lead the design and build-out of Temasek's enterprise semantic layer on Snowflake: the governed foundation that lets AI agents, BI tools and analysts work from consistent business definitions and trusted metrics. This covers how metrics, dimensions, entities and relationships are modelled, how semantic views are built and deployed, and how they are exposed to Cortex Agents and Snowflake Intelligence for reliable natural-language analytics.

You will work closely with business teams, data owners, governance, cybersecurity and enterprise architecture to make sure the semantic layer is accurate, secure and widely adopted. You stay current on where semantic and agentic AI capabilities are heading and help the broader organisation keep pace.

Semantic Layer Strategy and Architecture

Define the technical strategy and reference architecture for the Snowflake semantic layer, covering semantic view scoping by business domain, metric and dimension standards, relationships, and integration with BI, APIs and AI agents.

Establish standards for how semantic views are built, versioned, tested and promoted through environments (CI/CD, dbt integration, naming conventions, documentation).

Design how agents consume the semantic layer, including routing across multiple semantic views, combining structured (Cortex Analyst) and unstructured (Cortex Search) retrieval, and exposure via APIs or MCP-compatible interfaces.

Own the semantic layer roadmap, prioritising domains and use cases so that near-term delivery stays aligned with longer-term data platform direction.

Build and Delivery

Lead hands-on delivery of semantic views across priority domains, translating business logic and existing BI models (e.g., Qlik, Power BI) into governed, reusable metrics.

Build and maintain the data pipelines that feed the semantic layer, using Apache Airflow for orchestration and AWS services for ingestion, storage and integration with Snowflake.

Build and tune Cortex Agents on top of the semantic layer, including tool configuration, custom instructions and agent scoping.

Drive answer accuracy through verified query libraries, evaluation sets and iterative optimisation of semantic views.

Apply dimensional modelling best practices (Kimball, star schemas, conformed dimensions) across curated data layers, and optimise for performance and cost.

Explore AI-assisted tooling (e.g., Cortex Code, Semantic View Autopilot) to speed up creating and maintaining semantic models.

Lead and mentor a small squad of engineers, owning sprint planning and delivery.

Security, Governance and Observability

Partner with data governance, cybersecurity and enterprise architecture teams to embed RBAC, row and column-level security, lineage and audit logging into the semantic layer and agent access.

Establish metric ownership, certification and change-management processes so definitions stay trusted as the layer scales across business units.

Put monitoring in place for semantic view usage, agent behaviour, answer accuracy, data quality and cost.

Stakeholder and Cross-Functional Partnership

Work with business and investment teams to capture, reconcile and agree on metric definitions.

Act as the go-to person for teams building agents or analytics on the semantic layer, providing guidance on patterns, tooling and best practices.

Communicate design decisions and trade-offs clearly to both technical and non-technical stakeholders.

Trends and Innovation

Track developments in Snowflake's semantic, Cortex and agent capabilities and the wider semantic layer ecosystem (e.g., dbt Semantic Layer, ontologies and knowledge graphs), and turn them into actionable recommendations.

Evaluate emerging tools and patterns for relevance to Temasek's context and advise on when to adopt, monitor or defer.

Requirements

Experience

5–8 years in data engineering, analytics engineering or data architecture, including at least 2 years building semantic or metrics layers in production.

Snowflake: deep hands-on experience across data modelling, performance and cost tuning, RBAC and security, and Snowflake's AI features: semantic views or models, Cortex Analyst, Cortex Agents and Cortex Search. Exposure to Snowflake Intelligence is a plus.

Apache Airflow: hands-on experience designing, scheduling and monitoring production pipelines (DAG design, dependency management, alerting, retries).

AWS: working experience with core data services such as S3, IAM, Lambda and Glue, and with integrating AWS with Snowflake (e.g., external stages, Snowpipe, storage integrations).

Experience with dbt (including testing of metrics) and CI/CD for data assets.

Experience productionising LLM-based data applications, ideally text-to-SQL, RAG or agentic workflows, including evaluation and accuracy tuning.

Experience leading engineers and working across business, governance, security and architecture stakeholders.

Exposure to financial services or investment data (portfolio, valuations, performance metrics) is a strong plus

Skills and Attributes

Strong command of dimensional modelling, advanced SQL and semantic layer concepts (metrics, facts, dimensions, relationships, hierarchies).

Solid understanding of how LLMs and agents consume data: semantic metadata, tool use, retrieval and structured vs. unstructured access.

Proficiency in Python and SQL for data pipelines, automation and agent tooling.

Ability to define standards and make design decisions that balance rigour with practical usability.

Clear communicator who can work with engineers and business teams alike.

Intellectually curious, someone who actively follows developments in data and AI and brings insights back to the team.