Lead Product Manager, Data Platform (Measurement & BI)
Grab · Singapore
About this role
The Data Platform team at Grab is seeking a Lead Product Manager to oversee the Measurement & BI charter. This role focuses on evolving Grab's metrics platform to support both human users and AI agents, driving the company's data-as-a-growth-engine strategy. Responsibilities include defining product strategy, managing the metrics and semantic layers, and ensuring the platform scales to meet the needs of diverse stakeholders from data scientists to non-technical business users. You will work closely with engineering to design scalable data architectures and implement agentic BI solutions. Key requirements include at least 6 years of technical product management experience, with a minimum of 2 years specifically in data or AI platforms. Candidates must possess strong knowledge of metrics/semantic layers, dimensional modeling, and data architecture (e.g., Trino, StarRocks, Pinot). Experience shipping data products for AI agents, such as text-to-SQL or LLM-integrated semantic layers, is essential. This is an individual contributor role based onsite at Grab's Singapore headquarters, reporting to the Head of Product, Data and AI Platforms. Benefits include comprehensive medical and life insurance, GrabFlex benefits, and various leave policies including parental and volunteering leave.
For full details about this role, click “Apply on company site” to view the employer’s original posting.
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 the Metrics and Semantic Layer
The core mandate is to transition Grab's reporting onto a governed metrics and semantic model. You must ensure that data definitions are consistent across the organization, enabling both human users and AI agents to access reliable, standardized data for decision-making.
Scaling Agentic Business Intelligence
You are responsible for defining the product strategy for agentic BI, which involves building data products that LLMs can consume. This requires designing interfaces like text-to-SQL or semantic layers that allow AI agents to interact with the platform effectively.
Balancing Platform Feasibility and User Needs
You must bridge the gap between complex data architectures (Trino, StarRocks, Pinot) and diverse user needs. This involves challenging engineering designs on cost and feasibility while ensuring the platform remains intuitive for non-technical business users.
Likely challenges to think through
Defining the Boundary Between Platform and Application
A significant challenge is determining where the platform's responsibility ends and the application layer begins for agentic BI. Prepare by articulating a clear framework for how a platform should provide the necessary data primitives without becoming overly prescriptive or rigid for downstream applications.
Managing Diverse Stakeholder Requirements
The platform serves everyone from data scientists to non-technical business users. You will face the difficulty of prioritizing features that satisfy power users' need for flexibility while maintaining the simplicity required for broader organizational adoption. Prepare by demonstrating how you have successfully managed conflicting stakeholder roadmaps in previous data platform roles.
Technical Depth in Data Architecture
The role requires deep technical knowledge of dimensional modeling and query engines to effectively guide engineering. You may be challenged on your ability to evaluate the performance implications of different data architectures. Prepare by being ready to discuss specific trade-offs between latency, cost, and scalability in large-scale data environments.
03
Practice your answers
Technical practice
Designing for Agentic BI
What to coverExplain how you would architect a semantic layer to support both human-facing BI tools and AI agents. Discuss the trade-offs between latency, accuracy, and context-window constraints when serving LLMs via text-to-SQL or semantic APIs.
Metrics Layer Governance
What to coverDescribe your approach to migrating fragmented reporting into a centralized, governed metrics layer. Address how you handle conflicting definitions across business units and ensure consistency in dimensional modeling.
Data Architecture Trade-offs
What to coverCompare the use cases for OLAP engines like Trino, StarRocks, and Pinot in a high-scale environment. Suggest how you would evaluate which engine to prioritize based on query patterns, concurrency requirements, and cost efficiency.
Dimensional Modeling at Scale
What to coverDiscuss the challenges of applying Kimball dimensional modeling in a modern data lakehouse architecture. Explain how you ensure data quality and performance when dealing with complex DAG-based orchestration.
Optimizing for AI Consumption
What to coverWhat specific metadata or structural requirements are necessary for an LLM to effectively query a semantic layer? Suggest how you would design the feedback loop to improve agent performance over time.
Platform vs. Application Boundaries
What to coverHow do you determine where the platform's responsibility ends and the application layer begins in an agentic BI ecosystem? Provide examples of how you would enforce this boundary to prevent technical debt.
Data Discovery and Cataloging
What to coverHow would you improve data discoverability for non-technical users while maintaining strict governance? Suggest strategies for integrating cataloging tools into the daily workflow of data scientists and business analysts.
Performance Troubleshooting
What to coverIf a critical dashboard or AI agent query experiences a sudden latency spike, what is your systematic approach to identifying the bottleneck? Mention specific tools or metrics you would inspect across the data stack.
Behavioural practice
Managing Conflicting Stakeholder Needs
What to coverDescribe a situation where you had to balance the requirements of technical data scientists against the needs of non-technical business users. Explain how you prioritized features to ensure the platform remained scalable for both.
Driving Adoption of Data Platforms
What to coverProvide an example of how you successfully drove the adoption of a new data tool or platform. Focus on how you measured success and addressed resistance from teams accustomed to legacy workflows.
Navigating Ambiguity in Strategy
What to coverTell me about a time you had to define a product strategy in a highly ambiguous environment. How did you validate your assumptions and ensure alignment with the broader company data strategy?
Collaborating with Engineering
What to coverDescribe a time you challenged an engineering design decision based on feasibility or cost. How did you communicate your technical concerns while maintaining a strong, collaborative relationship with the team?
02
Sharpen your knowledge
Topics to prepare
Agentic Business Intelligence Architecture
Study the integration of LLMs with semantic layers to enable text-to-SQL and natural language querying. Exercise: Map out a conceptual architecture where an LLM agent interacts with a metrics layer, identifying how to handle context window limitations and query accuracy for non-technical users.
Metrics and Semantic Layer Governance
Understand the transition from traditional BI to governed metrics platforms. Exercise: Define a strategy for migrating ad-hoc reporting to a centralized semantic model, focusing on how to maintain consistency across diverse stakeholders like Finance and Data Science.
Data Platform Scalability and Performance
Review the performance characteristics of OLAP engines like Trino, StarRocks, and Pinot. Exercise: Analyze the trade-offs between these engines regarding latency and concurrency when serving both human dashboards and automated AI agent queries.
Skills to revise
Dimensional Modeling and Kimball Methodology
Apply Kimball dimensional modeling to complex data sets. Exercise: Design a star schema for a ride-hailing or delivery event stream, ensuring the model supports both historical analysis and real-time metric calculation.
Technical Product Strategy for Data Platforms
Practice defining product roadmaps for internal platform teams. Exercise: Draft a 6-month roadmap for a metrics platform, prioritizing features that balance immediate stakeholder needs with long-term architectural scalability.
Data Product Lifecycle Management
Manage the end-to-end delivery of data products. Exercise: Create a framework for measuring the success of a new data platform feature, specifically defining metrics for adoption, engagement, and task success for both human and AI users.
04
Your preparation checklist
Prepare a case study on semantic layer architecture
Draft a technical walkthrough of a project where you implemented a metrics or semantic layer (e.g., dbt, Cube, or LookML). Focus on how you handled dimensional modeling, ensured data consistency across stakeholders, and optimized performance for underlying engines like Trino or StarRocks.
Develop a framework for agentic BI integration
Prepare a brief proposal or design document explaining how you would architect a semantic layer to support LLM-based agents. Specifically, address how you would handle text-to-SQL challenges, ensure governance, and maintain accuracy when AI agents query your data platform.
Map your experience to Grab's data stack
Review your past projects to identify specific instances where you managed DAG-based orchestration or scaled data platforms. Be ready to discuss the trade-offs you made regarding cost, latency, and usability when working with high-volume data environments.
Formulate strategic questions for the Head of Product
Prepare thoughtful questions regarding the current maturity of Grab's metrics platform, the biggest technical bottlenecks in their existing data architecture, and how they define success for the transition toward agentic BI.
Refine your stakeholder management narrative
Prepare examples of how you have balanced the needs of technical users (data scientists, engineers) with non-technical business users. Highlight how you translated complex data platform requirements into clear product roadmaps and business value.
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?