Senior Associate/Assistant Vice President, AI Data Engineer

Temasek International Pte. Ltd.; Singapore; On-site; SGD 120,000 - 170,000; Full time

Job summary

Temasek International Pte. Ltd. seeks an experienced data architect to design and evolve an enterprise data platform that serves AI applications. You will shape architectures spanning APIs, vector databases, and graph stores to enable scalable, semantic access.

Own the data layer for RAG pipelines, establish ontology and schema standards, and define latency SLAs for real-time feeds. Collaborate with AI security, governance, and data teams to ensure compliant, well-documented data across domains.

Qualifications

• Experience designing data architectures optimized for AI/ML consumption.
• Experience implementing data quality gates and governance.
• Experience building end-to-end data pipelines with scalable storage and retrieval.

Responsibilities

• Design and build data architectures optimised for AI agent consumption, including structured stores exposed via APIs, vector databases for semantic retrieval, graph databases for relationship reasoning, and hybrid retrieval systems.
• Own the data layer for RAG pipelines: document ingestion workflows, chunking strategies, embedding generation and refresh, metadata tagging, and vector index management across domains.
• Establish ontology and schema standards to ensure AI-accessible data is consistent and easily interpretable.
• Architect real-time and near-real-time data feeds, defining latency and freshness SLAs.

Skills

Data architecture

Data governance

Data quality

Data lineage

AI data pipelines

Job description

Agent-ready data architecture

• Design and build data architectures optimised for AI agent consumption, including structured stores exposed via APIs, vector databases for semantic retrieval, graph databases for relationship reasoning, and hybrid retrieval systems combining keyword, semantic, and structured queries.
• Own the data layer for RAG pipelines: document ingestion workflows, chunking strategies, embedding generation and refresh, metadata tagging, and vector index management across domains (e.g., company research, market intelligence, portfolio data, regulatory filings).
• Establish ontology and schema standards to ensure AI-accessible data is consistent, well-documented, and interpretable without custom parsing logic.
• Architect real-time and near-real-time data feeds (e.g., market data, news, portfolio events), defining and enforcing latency and freshness SLAs.

Enterprise data quality and governance

• Define and implement data quality standards (completeness, consistency, freshness, anomaly detection) with automated quality gates to prevent degraded data entering AI systems.
• Build end-to-end data lineage tracking across AI pipelines, enabling traceability from source to AI consumption for debugging and audit requirements.
• Partner with AI Security & Governance and enterprise data teams to ensure compliance with data classification, access control, and cross-border handling requirements (including China-related workflows).
• Design and operate data observability tooling covering pipeline health, data drift, schema changes, and SLA monitoring, giving product teams visibility into data reliability.
• Run regular data quality reviews with AI product teams to identify gaps impacting performance and prioritise data engineering investments.

Shared, reusable data platform for AI

• Develop reusable data assets and services supporting multiple AI products, including a shared investment knowledge graph, company/market data APIs, document intelligence pipelines, and portfolio analytics services.
• Maintain a data catalogue documenting sources, schemas, freshness, quality metrics, access protocols, and limitations to enable informed data usage.
• Contribute to enterprise data platform strategy with an AI-first perspective, ensuring architectures support AI consumption patterns beyond traditional BI/reporting needs.
• Engage external data vendors to evaluate, onboard, and maintain high-quality data sources, including ongoing quality assessment and licensing management with procurement.