Senior Data Agent Engineer (Data Intelligence)

Key Responsibilities

• Own the end-to-end technical design and delivery of AI-powered business analytics applications, enabling a complete analytical loop from conversational data querying to anomaly attribution, insight generation and decision recommendations.
• Lead the design and implementation of the intelligent attribution engine, covering dimensional attribution (multi-dimensional drill-down with contribution calculation), factor attribution (metric calculation-tree decomposition) and correlation-based attribution, to automatically identify root causes of key metric fluctuations.
• Design automated metric monitoring and anomaly detection mechanisms, enabling AI to proactively surface issues and push attribution findings to stakeholders.
• Lead the build-out of the metrics semantic layer, defining machine-readable business definitions, calculation logic, analysable dimensions and synonyms, so that natural-language intent is accurately translated into structured queries.
• Package data and analytical capabilities into standardised tool interfaces (Function Calling / MCP protocol) for consumption by AI applications.
• Build and maintain business analytics data models, taking ownership of dimensional modelling, metric processing and data quality assurance.
• Establish evaluation and observability frameworks to quantify query accuracy, metric-definition correctness and reliability of attribution conclusions, driving a closed-loop bad-case optimisation process.
• Implement controls over AI data access, including permission boundaries, data masking, data lineage, result explainability and full audit trails, in line with Hong Kong licensed corporation regulatory requirements.
• Define data and AI technical standards, mentor junior team members, and drive cross-departmental alignment on metric definitions.

Requirements

• Bachelor’s degree or above in Computer Science, Data, Statistics or a related discipline; minimum 8 years of experience in data engineering or data intelligence, including at least 2 years leading projects independently.
• Expert-level SQL (complex queries, window functions, performance tuning) and strong proficiency in Python.
• Deep expertise in dimensional modelling and data warehouse layering (ODS/DWD/DWS/ADS), with proven experience independently leading a complete data domain or data platform build.
• Solid LLM application engineering capability, with in-depth understanding of RAG, Function Calling / Tool Calling and Text-to-SQL — including their implementation approaches, capability boundaries and failure modes — and the ability to constrain probabilistic generation with deterministic rules.
• Hands-on experience with at least one data warehouse or compute engine (Hive, MaxCompute, Greenplum, ClickHouse, Databricks, etc.) and orchestration.