17 days ago
Shenzhen, ChinaStaff+
Responsibilities
- Build and operate pipelines for structured domain data including onboarding, customer/entity, credit, risk, portfolio, performance, and arrears data.
- Build unstructured document pipelines covering parsing, metadata enrichment, deduplication, and retention handling.
- Develop embedding and vectorisation pipelines and manage vector indices with governed refresh and deletion strategies.
- Implement automated testing, service-level monitoring, anomaly detection, observability, data lineage, and operational runbooks.
- Apply privacy-by-design through PII handling, masking or tokenisation, access controls, and audit trails.
- Implement dataset versioning, snapshots, schema evolution, data contracts, and reproducible evaluation workflows.
- Partner with Agent Engineers and stakeholders to define data requirements and evaluation datasets.
- Support production operations and continuous improvement of AI data services.
- Improve retrieval relevance and answer grounding through metadata, input chunking, and data quality improvements.
Requirements
- Experience delivering reliable, fresh, well-governed datasets and indices for AI solutions.
- Experience implementing controlled access patterns for sensitive data domains such as KYC/AML or credit risk is preferred.
- Familiarity with vector search concepts, document ranking or reranking, and retrieval metrics is preferred.
- Experience with feature stores or MLOps-style dataset management is preferred.
Benefits
- Flexible working, continuous professional development, and opportunities for career growth within an inclusive and diverse workplace.
Categories
Data Engineering
About HSBC
HSBC is a global bank that provides retail, commercial and investment banking, payments, and wealth management to individuals, SMEs, and multinationals. It earns fees, interest income, and trading revenues through branches and digital platforms across major markets. Headquartered in London, it serves over 40 million customers in 58 countries and is listed in London, Hong Kong, New York, and Bermuda.