Business Intelligence Engineer, Marketplace Productivity Solutions

Description

Japan Seller Services is seeking a Business Intelligence Engineer to join the Marketplace Productivity Solutions team. This is not a traditional BI role focused on dashboards for leadership. You will be the person who designs the analytical intelligence inside an AI-powered platform — determining what insights get generated, how seller data is structured for AI consumption, and whether the output actually drives Account Manager action.

Our platform automatically generates seller performance reports with AI-powered insights and recommendations for 270+ Account Managers. The quality of those insights depends on someone who deeply understands the data: which metrics combinations surface actionable signals, what “decline root cause” logic looks like in SQL, how to structure data context so an LLM produces relevant recommendations rather than generic summaries. That person is you.

You will use SQL and Python daily — not just to pull data, but to define the analytical logic that powers automated reports, design the data payloads that feed AI models, evaluate whether AI-generated insights are accurate, and continuously expand the platform’s analytical capabilities. You’ll work at the intersection of traditional BI (metrics, benchmarks, comparisons) and generative AI (prompt context design, output evaluation, quality iteration), making this a uniquely forward-looking BIE role.

Key job responsibilities

• Insight Design

Define the analytical logic (SQL/Python) that powers automated seller reports: YoY/MoM/WoW comparisons, benchmark calculations, contribution-to-change analysis, decline root cause identification, and opportunity signals.

Design and maintain the metric frameworks behind each report type — determining which KPIs, breakdowns (GL/Node/Brand/ASIN), and comparisons are most relevant for each user persona.

Develop reusable analytical models that identify seller-level patterns: growth acceleration, underperformance vs. peers, untapped program eligibility, and seasonal anomalies.

Collaborate with Product Managers to translate user feedback into new analytical angles and measurable improvements.

• AI Insight Quality

Design and refine the data context passed to LLMs for insight generation — determining which metrics, comparisons, and seller attributes produce the most relevant outputs, while keeping the payload information-dense and concise.

Evaluate LLM-generated insights for accuracy, relevance, and actionability — ensuring AI recommendations are grounded in real seller performance data, not hallucinated.

Build automated evaluation frameworks to measure insight quality at scale (accuracy rates, recommendation specificity, user satisfaction correlation) and support continuous quality monitoring.

Identify where AI output is weak or generic, and drive improvements through data-layer adjustments, context optimization, or prompt refinements in partnership with Data Scientists.

• Report & Template Expansion

Continuously expand the report library — designing the data logic for new report types covering different growth levers, user personas, and analysis cadences (monthly, weekly, custom date range).

Build and validate new metric calculations as business needs evolve — ensuring consistency with globally standardized definitions while accommodating JP-specific requirements.

Define the data requirements and analytical logic for new platform features, and measure adoption and impact to inform prioritization of future report and template investments.

Basic Qualifications

• 3+ years of experience writing complex SQL queries against large-scale data warehouses (Redshift, Athena, or equivalent)
• Proficiency in Python for data processing, analysis automation, and building analytical pipelines
• Experience designing metrics frameworks and translating business questions into structured analytical approaches
• Experience with data visualization tools (QuickSight, Tableau, or similar) for stakeholder-facing reporting
• Ability to translate ambiguous business problems into clear data requirements and actionable analysis
• Business-level Japanese (N1 or native) and business-level English (written and verbal) — required for collaboration with JP business stakeholders and global technical teams
• Bachelor’s degree or above in computer science, data science, statistics, engineering, economics, or related quantitative field

Preferred Qualifications

• Experience working with generative AI outputs — evaluating LLM quality, designing prompts with data context, or building automated evaluation pipelines
• Experience optimizing data inputs for AI/ML models (feature engineering, context design, signal-to-noise optimization)
• Experience building data pipelines and ETL processes in AWS (Lambda, S3, Athena, Step Functions)
• Experience with statistical analysis and experimentation (A/B testing, causal inference, impact measurement)
• Background in e-commerce, marketplace, or sales operations analytics
• Experience with data modeling, warehousing, and building scalable analytical infrastructure
• Familiarity with AI-assisted development tools (Claude Code, GitHub Copilot) to accelerate analytical workflows
• Master’s degree in a quantitative field

Our inclusive culture empowers Amazonians to deliver the best results for our customers. If you have a disability and need a workplace accommodation or adjustment during the application and hiring process, including support for the interview or onboarding process, please visit https://amazon.jobs/content/en/how-we-hire/accommodations for more information. If the country/region you’re applying in isn’t listed, please contact your Recruiting Partner.

Company - Amazon Japan G.K. - A43

Job ID: A10466322