Data Scientist - Agent Experience Department (AGEX)

Job Description:

Business Overview
Rakuten is one of the biggest marketplace of Japan and is the largest internet ecosystem with a wide range of services ranging from e-Commerce, Travel, Banking, Fintech, Food Delivery, Golf, Insurance, Instant Messaging, Mobile Network etc. Rakuten has over 140 services globally. Our mission is to empower people and society through the internet while aiming at becoming the Global Innovation Company.

Department Overview

The Agent Experience (AGEX) team is pioneering the development of Rakuten AI Companion and AI Composer: two complementary AI initiatives designed to transform how users engage with the Rakuten ecosystem and beyond.
The Rakuten AI Companion is a next-generation personal assistant that supports a wide range of user needs, from navigating Rakuten’s services to assisting with everyday goals. Built with a user-centric and task-oriented approach, it delivers an experience that is intuitive, helpful, and truly personal: empowering users across all aspects of their digital lives.
Running in parallel, the AI Composer is engineered to infuse seamless AI-driven experiences across Rakuten’s product ecosystem: generally, contextually, and personally.

By embedding intelligence into user touchpoints, AI Composer ensures that every interaction feels natural, anticipatory, and deeply integrated with individual needs.
Together, AI Companion and AI Composer form the foundation of Rakuten’s next-generation AI Agent Experience: intelligent, adaptive, and human-centered.

Position:
Why We Hire
We’re seeking a data scientist to join the Agent Science team and play a key role in building and improving our AI companion. You’ll operate at the intersection of data science, product insight, and applied research: analyzing early user signals, uncovering system limitations, and driving improvements that shape assistant quality and user value. You’ll also help inform the product roadmap through insights derived from user behavior and the broader Rakuten ecosystem.

Position Details

Responsibilities
- Analyze assistant interactions and internal evaluations to identify improvement opportunities.
- Generate data-driven insights to guide product development and feature prioritization.
- Conduct research to improve agent architecture and user intent understanding to enhance response quality.
- Design and execute experiments, identifying the right metrics to evaluate impact.

- Collaborate with engineers and other Agent Science team members to implement and productionize improvements.
- Leverage diverse datasets across Rakuten’s ecosystem to improve assistant capabilities and deliver customer value.
- Clearly communicate findings and contribute to research and product decisions.
 

Mandatory Qualifications:

- 3+ years of experience in data science, machine learning, or a related quantitative field.

- Strong Python and ML skills for data processing, analysis, and prototyping.

- Proficiency in SQL and working with large-scale structured data.

- Solid foundation in statistics, experimental design, and predictive modeling.

- Possesses strong problem identification and formulation skills, with a track record of proactively identifying potential issues and developing effective solutions.

- Demonstrated ability to thoroughly research and evaluate potential solutions and identify the optimal approach for a given problem.

- Excellent written and verbal communication skills, with the ability to clearly articulate complex technical concepts to both technical and non-technical audiences.

Desired Qualifications:

- Experience with conversational AI, NLP, or Agentic applications.

- Familiarity with LLM evaluation methods and user interaction data.

- Experience working in product-oriented, cross-functional environments.
- Basic Japanese language skills.

Other Information:

Additional information on Location

Rakuten Crimson House (Head Office)

#engineer #datascientist #technologyservicediv #Python