The Data Platform Principal Architect will join the Knowledge Platform team at Airwallex, serving as the primary technical authority for the company's data and AI strategy. This role focuses on evolving the data ecosystem into an AI agent-ready infrastructure, bridging platform architecture, customer-facing data products, and AI enablement. Responsibilities include designing governed, well-modeled structured datasets, architecting a semantic layer for human and machine consumption, and defining operationalized capabilities for AI agents. The architect will lead global data modeling efforts, implement Governance-as-Code, and ensure regional data localization strategies are executed within the Databricks environment. Required qualifications include 10+ years of experience in large-scale data systems, hands-on expertise with the Databricks ecosystem (Unity Catalog, Delta Lake), experience with semantic layers or metric stores, and a solid understanding of AI/agent-based systems. Candidates must demonstrate senior-level individual contributor experience, including leading design reviews and mentoring engineers. The role is based in Singapore.
For full details about this role, click “Apply on company site” to view the employer’s original posting.
Skills for this role
Data ArchitectureDatabricksUnity CatalogDelta LakeSparkKafkaSemantic LayerMetric StoresData ModelingAPI DevelopmentGovernance-as-CodeData ResidencyComplianceAI AgentsRAG
Your skill match
Checking your profile…
YOUR NEXT STEP
Get interview-ready for this role
A focused preparation guide, built around this job’s responsibilities and requirements.
✦ AI-generated guide
Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.
01
Understand the role
What matters most
Evolving Data Infrastructure for AI Agents
The primary mandate is to transition the existing data ecosystem into an AI agent-ready infrastructure. You must design the 'Skills Layer'—the architecture for agent-callable tools, analytical workflows, and transformation primitives—that allows AI agents to perform reliable reasoning and actions rather than simple data retrieval.
Architecting a Unified Semantic Layer
You are responsible for building a semantic layer that provides consistent business metrics and entity relationships. This is critical for ensuring that both human analysts and AI agents receive trustworthy, uniform answers to complex business questions across different regions without needing to reverse-engineer underlying schemas.
Balancing Global Consistency with Regional Compliance
You must design scalable data patterns that maintain a unified global model while strictly adhering to regional data residency and localization requirements. This requires reconciling local compliance constraints with global consistency in definitions and metrics within the Databricks environment.
Implementing Governance-as-Code
You must embed automated data lineage, quality controls, and governance directly into the platform architecture. This ensures that as the system scales to support AI agents, the data remains clean, trusted, and compliant by design rather than through manual oversight.
Likely challenges to think through
Reconciling Global Models with Localized Data Residency
The difficulty lies in creating a unified global schema that satisfies strict regional regulatory requirements. Prepare by demonstrating how you have previously designed architectures that decouple logical global modeling from physical data storage to meet compliance without sacrificing analytical utility.
Designing Reliable Interfaces for AI Agents
AI agents require highly deterministic and reliable tool interfaces to function correctly. The challenge is moving beyond standard API design to create 'Agent Interfaces' that provide sufficient context and safety. Prepare by discussing your experience in building robust, agent-callable services that handle error states and data provenance effectively.
Managing Technical Debt During Rapid Platform Evolution
Evolving an existing data ecosystem into an AI-native one while maintaining current operations is complex. Prepare to discuss how you prioritize architectural refactoring versus new feature development, specifically how you maintain system stability while implementing significant changes to the data and knowledge layers.
03
Practice your answers
Technical practice
Architecting for AI Agent Readiness
What to coverExplain how you would evolve a traditional data platform into an AI agent-ready infrastructure. Focus on the design of 'Agent Interfaces' and how to expose data, knowledge, and skills to LLMs while maintaining security and performance.
Semantic Layer Implementation
What to coverDescribe your approach to building a semantic layer that serves both human analysts and AI agents. Discuss how you ensure consistent metric definitions and entity relationships across diverse business domains.
Governance-as-Code in Databricks
What to coverHow would you implement Governance-as-Code within the Databricks Unity Catalog environment? Detail your strategy for automating data lineage, quality controls, and access policies as part of the CI/CD pipeline.
Regional Data Localization vs. Global Consistency
What to coverDiscuss the trade-offs between maintaining regional data residency compliance and achieving a unified global data model. Suggest architectural patterns to reconcile these requirements without sacrificing data utility.
Designing for Real-Time Decision Systems
What to coverWhat architectural patterns do you use to support real-time performance for data-driven decision-making? Address the integration of streaming data (e.g., Kafka) with batch processing in Delta Lake.
Data Modeling for AI/ML Workflows
What to coverHow do you design structured datasets to support both traditional BI and AI/ML feature engineering? Explain your approach to balancing normalization for transactional integrity with denormalization for model performance.
Scaling Data Infrastructure
What to coverGiven your 10+ years of experience, describe a scenario where you had to re-architect a large-scale data system to handle significant growth. Focus on the technical decisions made to improve scalability and maintainability.
API Development for Data Services
What to coverWhat are the key considerations when building APIs that allow AI agents to 'act' on data rather than just querying it? Discuss security, rate limiting, and state management for agent-callable tools.
Behavioural practice
Leading Through Influence and Mentorship
What to coverDescribe a time you led a complex design review or mentored engineers to improve their architectural decision-making. Focus on how you balanced your role as a technical authority with the need to foster team growth.
Navigating Conflicting Stakeholder Requirements
What to coverProvide an example of a situation where you had to reconcile conflicting requirements between Product, Risk, and Engineering teams. Explain how you reached a consensus while ensuring the technical integrity of the platform.
Making Decisions from First Principles
What to coverShare a specific instance where you had to make a high-stakes architectural decision based on first principles rather than following industry trends. What was the situation, and what was the long-term impact of your decision?
Handling Regulatory and Compliance Challenges
What to coverDescribe a project where you had to design a data solution under strict regulatory constraints. Detail the situation, the specific actions you took to ensure compliance, and the final result for the organization.
02
Sharpen your knowledge
Topics to prepare
AI Agent-Ready Data Architecture
Review the conceptual shift from traditional data warehousing to agent-ready infrastructure. Exercise: Draft a high-level design document explaining how to expose a semantic layer to LLM-based agents via APIs, focusing on tool-calling capabilities and context management.
Governance-as-Code and Data Compliance
Study the integration of automated policy enforcement within data pipelines. Exercise: Outline a strategy for implementing automated lineage and quality checks in a multi-region environment that satisfies strict data residency requirements.
Semantic Layer Design
Focus on creating a unified metric store that serves both human analysts and machine agents. Exercise: Define a schema strategy that reconciles global metric consistency with regional variations in business logic.
Skills to revise
Databricks Unity Catalog and Delta Lake
Master the technical implementation of Unity Catalog for fine-grained access control and lineage. Exercise: Configure a sample catalog structure that enforces regional data isolation while maintaining global visibility for authorized roles.
API Development for AI Agents
Develop expertise in building interfaces that allow AI agents to interact with data platforms. Exercise: Design a set of REST or GraphQL endpoints that enable an agent to query structured datasets and retrieve metadata from a semantic layer.
04
Your preparation checklist
Prepare a Databricks Unity Catalog architectural case study
Develop a concise walkthrough of a large-scale data system you designed using Unity Catalog and Delta Lake. Focus on how you managed data lineage, access control, and multi-region data residency requirements within a unified global model.
Draft a framework for Governance-as-Code
Outline your approach to implementing automated data quality and lineage controls. Be prepared to discuss how you integrate these checks into CI/CD pipelines to ensure compliance and consistency across distributed data environments.
Design a semantic layer for AI agents
Create a conceptual model for a semantic layer that supports both human analysts and AI agents. Explain how you would structure metric definitions and entity relationships to ensure consistent, trustworthy outputs for natural language querying.
Formulate strategies for AI agent-ready infrastructure
Prepare to discuss the 'Skills Layer' architecture. Specifically, define how you would build agent-callable tools and APIs that allow AI systems to reliably reason over and act upon structured data.
Develop technical questions for the Knowledge Platform team
Prepare thoughtful questions regarding their current challenges with regional data localization, the specific AI agent use cases they are prioritizing for 2026, and how they balance global schema consistency with local regulatory constraints.
Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy
Share this opportunityKnow someone who would be a great fit?