New opportunity

Manager/Senior Mgr Data & AI Product Mgr, Data Exploration & Partnerships

STB Singapore Tourism Board · Singapore

About this role

The Data and AI Product Manager will develop data and AI products for Singapore Tourism Board (STB) and tourism industry stakeholders. The role supports Stan, the Singapore Tourism Analytics Network, which provides tourism data and insights to internal and external users. Based in the Data Exploration and Partnerships department and reporting to its head, the manager works with STB units, government agencies and industry partners to improve tourism decision-making, user adoption and measurable business outcomes.

Responsibilities

Develop and execute the data product roadmap; identify stakeholder needs and new product opportunities; write requirements, acceptance criteria and test cases; and create prototype dashboards and datasets. Work with the Stan team on implementation, testing, production verification, launches and go-to-market plans.
Shape and prioritise an AI roadmap for STB data platforms based on feasibility, impact and readiness. Develop hands-on proofs of concept, beginning with natural language query for conversational access to datasets and analytics. Define AI evaluation metrics, test cases and guardrails, and document proofs of concept for handover to implementation teams.
Map user journeys and business scenarios, translate needs into product specifications, and improve processes for implementing and maintaining products. Use tools such as FormSG and Plumber where appropriate to automate data collection and basic validation.
Support data governance, security, compliance and quality for industry-facing data; coordinate stakeholders with varying technical maturity; and support data pilots involving analytics, integration and wrangling. Track product KPIs, usage, feedback and performance to guide improvements.

Qualifications

Formal training in Business, Data Science, Information Systems, Computer Science, Engineering or a related field, with at least 3–5 years of experience in data product management, data engineering or related roles and a record of delivering data-driven solutions.
Strong analytics, data exploration, transformation and dashboard skills, including AWS QuickSight; SQL proficiency and at least one of Python or R. Hands-on AI prototyping, prompt design, AI output evaluation and simple workflows; familiarity with data modelling, governance and quality. Product management, critical thinking, communication and independent execution are important.
Advantageous: tourism, hospitality or public-sector experience; cloud analytics platforms, data integration, API management, AI application development, prompt engineering or LLM evaluation. Relevant product management, data engineering or AI/ML certifications are welcome.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

AI product managementData product managementProduct roadmappingRequirements gatheringUX designGo-to-market planningStakeholder managementData analyticsData explorationData transformationDashboard developmentAWS QuickSightSQLPythonRAI prototypingNatural language queryPrompt designAI output evaluationLLM evaluationData modellingData integrationData wranglingData governanceData quality managementRegulatory complianceAPI managementFormSGPlumber

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Take data products from roadmap to launch
The role owns the data product roadmap, specifications, acceptance criteria and test cases. It also develops mock dashboards or datasets, works with the Stan team on implementation and production verification, and leads go-to-market plans for major releases.
Prove AI use cases before scaling them
STB calls for an AI roadmap prioritised by feasibility, impact and readiness, starting with a hands-on natural-language-query proof of concept. Evaluation metrics, test cases, guardrails and documentation for implementation handover are central to the work.
Design for reliable industry-facing data
The role supports data integrity, security, compliance and quality during productisation. It must also apply governance and quality criteria to AI outputs, not just to underlying datasets.
Align diverse stakeholders around usable solutions
The manager works with STB units, government agencies and industry partners with varying technical maturity. User journeys, clear problem statements and actionable specifications must connect their needs with what the Data Partnerships and Stan teams can implement.
Demonstrate adoption and improve outcomes
Success is described in terms of better tourism decision-making, user adoption, stakeholder satisfaction and measurable impact. The role defines product KPIs and uses usage patterns, feedback and performance data to guide improvements.

Likely challenges to think through

Reconciling competing product needs
Likely challenge, not a confirmed problem: internal and industry users may value different data products, while implementation capacity and technical feasibility constrain the roadmap. Prepare a prioritisation approach that makes user value, organisational goals, feasibility and trade-offs explicit.
Moving conversational analytics beyond a demo
Likely challenge, not a confirmed problem: a natural-language-query proof of concept may appear useful yet produce inaccurate or inappropriate answers or be difficult to hand over. Prepare to explain how you would test representative queries, set evaluation criteria and guardrails, and document what an implementation team needs.
Balancing access with data safeguards
Likely challenge, not a confirmed problem: making tourism insights useful to external stakeholders can pull against data security, compliance and quality needs. Prepare a product-design approach that checks data reliability and governance requirements before release, then monitors feedback and performance afterward.

Practice your answers

Technical practice

How would you prioritise a data product roadmap for Stan's internal and industry users?

What to coverThe posting calls for a roadmap aligned with STB's goals and stakeholder needs. Preparation suggestion: explain how you would compare decision-making value, user demand, feasibility and dependencies, then revisit priorities as evidence changes.

How would you turn a tourism analytics need into testable product requirements?

What to coverThe role includes user journeys, specifications, acceptance criteria, test cases and prototype dashboards or datasets. Preparation suggestion: walk through a user scenario, define the expected insight and data, and show how a prototype and acceptance tests would resolve ambiguity before implementation.

How would you explore and combine datasets for a data pilot?

What to coverThe posting names SQL, Python or R, data integration, wrangling and analytics as relevant skills. Preparation suggestion: discuss joins and data grain, transformation checks, missing or inconsistent values, and how you would verify that the resulting dataset answers the pilot's question.

How would you design a dashboard that helps users make tourism decisions?

What to coverThe role involves prototype dashboards, data exploration and dashboard development using tools such as AWS QuickSight. Preparation suggestion: explain how you would choose measures and visualisations for a defined user journey, test whether users interpret them correctly, and balance useful detail against clarity.

How would you prototype natural-language queries over Stan datasets?

What to coverThe posting identifies conversational access to datasets and analytics as the starting AI proof of concept. Preparation suggestion: cover dataset definitions, modelling and prompt design, then explain how you would handle ambiguous questions and check answers against the underlying data.

How would you decide whether an AI analytics feature is ready for productisation?

What to coverThe role calls for AI evaluation metrics, test cases and guardrails covering reliability, accuracy and appropriateness. Preparation suggestion: define representative and failure-case queries, describe how outputs would be checked, and distinguish a promising demonstration from a feature safe to hand over.

How would you protect the quality and appropriate use of industry-facing data?

What to coverThe posting assigns responsibility for data integrity, security, compliance and quality during productisation. Preparation suggestion: discuss validation, access and output controls, ownership of quality issues, and the trade-off between data usefulness and responsible disclosure without assuming a particular STB policy.

Where would automation help in data collection and validation?

What to coverThe posting cites FormSG and Plumber as possible tools for data collection and simple validation checks. Preparation suggestion: describe a bounded workflow, identify checks suited to automation, and explain when an exception should receive human review.

What would you need to move a data or AI pilot into production on Stan?

What to coverThe role spans pilots, implementation with the Stan team, testing, production verification and documented AI proof-of-concept handover. Preparation suggestion: explain how you would assess scalability and operational readiness, capture specifications and test evidence, and resolve gaps exposed by the pilot.

How would you evaluate a major data product release after launch?

What to coverThe posting includes go-to-market plans and tracking product KPIs, usage, feedback and performance. Preparation suggestion: propose measures tied to adoption, stakeholder satisfaction and improved decision-making, then explain how you would use weak or conflicting signals to guide the next release.

Behavioural practice

Tell me about a time you brought structure to an ambiguous product problem.

What to coverThe posting emphasises breaking down uncertain problems and facilitating decisions. Preparation suggestion: use a real situation, your actions to establish a problem statement and next steps, and the resulting decision or outcome.

Describe a time you aligned stakeholders with different data or technical maturity.

What to coverThe role works across STB units, government agencies and industry partners with varying levels of maturity. Preparation suggestion: give the situation, how you adapted discussions and reconciled needs, and the observable result without implying those were your past stakeholder groups.

Tell me about a time you explained a complex data or AI trade-off to a non-technical audience.

What to coverThe posting calls for clear communication that enables understanding and buy-in. Preparation suggestion: identify the decision at stake, describe how you made the trade-off understandable, and state what the audience decided or did.

Describe a time you took a pilot from exploration toward an implementable outcome.

What to coverThe role balances hands-on proof of concept work with documented handover and follow-through. Preparation suggestion: explain the situation, what you personally tested and documented, and the result, including any limits the pilot revealed.

Tell me about a time feedback or usage evidence changed your product plan.

What to coverThe posting expects continuous improvement based on usage patterns, stakeholder feedback and performance metrics. Preparation suggestion: set out the initial plan, the evidence and action you took, and the effect on users or the product.

Sharpen your knowledge

Topics to prepare

Data and AI roadmap prioritisation
Exercise: Rank three hypothetical tourism-data features by stakeholder value, feasibility and implementation readiness, then explain the trade-offs. Relevance: The posting calls for a data product roadmap and an AI roadmap prioritised by feasibility, impact and readiness.
Concept-to-production productisation
Exercise: Outline decision gates from prototype to pilot, testing, production verification, launch and subsequent improvement. Relevance: The role spans hands-on proofs of concept, handover to implementation teams, go-to-market planning and evaluation of pilots for scalability.
AI evaluation for natural language query
Exercise: Define test cases for a hypothetical conversational tourism-data query, including an ambiguous question, an incorrect answer and a request for restricted data; specify accuracy checks and guardrails. Relevance: Natural language query is the stated first AI proof of concept, and the posting requires evaluation metrics, test cases and guardrails.
Data modelling, governance and quality
Exercise: Sketch how a hypothetical visitor dataset’s fields and relationships support a dashboard and natural language queries; identify checks for missing values, accuracy and appropriate access. Relevance: The posting calls for modelling to support AI use cases and governance, security, compliance and quality for industry-facing data.
Stakeholder outcomes and product KPIs
Exercise: Map the different needs of an internal STB user and an industry stakeholder, then propose measures of usage, satisfaction and decision-making value. Relevance: Stan serves both audiences, and the role tracks adoption, feedback, performance and measurable business impact.

Skills to revise

SQL and Python or R analysis
Exercise: Using a hypothetical tourism dataset, write SQL to join, filter and aggregate records; use Python or R to inspect anomalies and transform fields for analysis. Relevance: The posting explicitly asks for SQL, at least one of Python or R, and data exploration and transformation.
AWS QuickSight dashboard prototyping
Exercise: Build a mock tourism-insights dashboard in AWS QuickSight using sample data, then write two acceptance checks for its figures and filters. Relevance: The posting names QuickSight proficiency and asks for mock or proof-of-concept dashboards and datasets.
Natural language query prototyping
Exercise: With synthetic data, prototype a short question-to-answer workflow, revise its prompt for ambiguous questions, and record expected versus actual outputs and failure cases. Relevance: The role starts AI proofs of concept with conversational dataset access and calls for prompt design, simple workflows and output evaluation.
Requirements and user-journey specification
Exercise: Turn a hypothetical partner request for tourism insights into a user journey, problem statement, product specification, acceptance criteria and test cases. Relevance: These are explicit responsibilities for designing data and AI products with stakeholders.
Stakeholder facilitation and handover
Exercise: Practise a short discussion that reconciles a non-technical stakeholder’s desired insight with an implementation team’s feasibility concern; document the decision, open questions and proof-of-concept handover notes. Relevance: The posting emphasises alignment across differing technical maturity levels and specifications suitable for productisation.
Data-collection process design
Exercise: As a preparation suggestion, diagram a sample collection process using FormSG and Plumber, with a simple required-field validation and an exception path; no access to STB systems is assumed. Relevance: The posting names these tools as possible ways to automate collection and basic validation where appropriate.

Your preparation checklist

Prepare a product delivery walkthrough
Choose a real data product example, if available, and outline the problem, stakeholder decisions, your contribution, pilot findings, production handover and results. The posting asks for delivery from concept to pilot to production and measurable user value.
Draft a Stan product design exercise
Using a clearly hypothetical tourism use case, produce a one-page user journey, prioritised roadmap item, requirements, acceptance criteria and test cases. Add a mock dashboard or dataset and explain how you would check feasibility with the Stan and Data Partnerships teams.
Assemble hands-on analytics evidence
Prepare a concise example of data exploration and transformation using SQL and Python or R, plus a dashboard walkthrough using AWS QuickSight if you have one. Explain the data model, validation checks and decisions the output supported; do not imply use of tools you have not used.
Sketch a natural-language-query proof of concept
Create a hypothetical conversational query flow over a dataset, with sample prompts, expected answers, failure cases and evaluation metrics. Show how you would assess feasibility, impact and readiness, apply guardrails, and document the proof of concept for implementation handover.
Map governance and operating controls
Draft a checklist for data integrity, access, security, compliance and quality checks before an industry-facing release. Include where simple collection and validation automation might help; the posting names FormSG and Plumber as examples, not mandatory tools.
Plan alignment, launch and measurement
Prepare a stakeholder discussion plan for STB units, government agencies and industry partners with differing technical maturity. Add a sample launch approach and KPIs for adoption, satisfaction and decision-making impact, including how usage and feedback would guide revisions.
Write targeted questions for STB
Ask which Stan user needs are most urgent, what datasets and constraints shape the first natural-language-query proof of concept, how pilot-to-production decisions are made, and which outcomes the department would measure first.

Availability last checked: 4 Nov 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…