My Rakuten Experience: Connie
Connie is a Product Manager in the AI Research Supervisory Department
Current Responsibilities
Currently, I am involved in driving key initiatives related to in-house Large Language Model (LLM) development, GPU infrastructure, and advanced platform tools. My main mission is to deliver scalable AI solutions through efficient resource management and user-centric product design.
Why Rakuten?
I have gained experience in a wide range of fields, including product management, user experience, and technology strategy. I have consistently focused on transforming complex technologies into tangible value, always with innovation and execution as my core principles.
I decided to join Rakuten because I felt it was an unparalleled opportunity to fully leverage my experience in the domain of AI and high-performance computing.
Job Satisfaction
One of the most rewarding moments in my career was being able to be involved in the LLM development project from its early stages. When this ambitious project began, our internal know-how regarding large-scale model training, especially complex tasks such as distributed training and large-scale optimization, was not yet sufficient. It was truly a challenge to build everything from scratch.
Together with a small cross-functional team, we boldly ventured into uncharted territory. We continuously learned through trial and error, steadily building the technical and operational foundations necessary to train world-class LLMs. Witnessing both the product and the team grow from initial uncertainties into a high-performing, self-sufficient AI capability has been incredibly fulfilling. This process of exploration and growth itself has been a great joy for me.
Attractions and Charms of Working at Rakuten
For me, the greatest attraction of working at Rakuten is the convergence of its business scale, future outlook, and individual ownership. Rakuten’s environment is truly unique -I can engage in innovative technologies such as LLMs and cutting-edge AI infrastructure, while having the autonomy to consistently lead from strategy formulation to execution.
Because of Rakuten’s diverse businesses, such as e-commerce, fintech, mobile, and digital content, our work in AI has the potential to bring significant impact across the entire organization and create value across various business domains. Knowing that our efforts contribute not only to Rakuten’s growth but also to innovation in Japanese society as a whole keeps me highly motivated.
All of this is possible because Rakuten’s corporate culture encourages experimentation, promotes continuous learning, and values collaboration with talented and motivated colleagues, which allows individuals and ideas to flourish.
Innovation
One of the projects I was involved in that had a particularly significant impact was the development of a semantic search model (*). This project started with no existing foundation, and we built everything from scratch. Through close collaboration between engineers and the business team, we successfully developed a scalable, AI-driven search solution that more deeply understands user intent.
Currently, this innovative solution has been implemented by over 10 business clients and has contributed to an increase in Gross Merchandise Sales (GMS) exceeding 1 billion JPY annually. It brings me a huge sense of accomplishment to see how our AI solution has driven both improved user experience and concrete business value on a large scale. I believe this project demonstrates Rakuten’s commitment to leveraging technology for real-world impact.
Lifestyle
I don’t strictly separate my work and private life. For me, personal learning and research are an indispensable part of my lifestyle, and they are closely intertwined with my work. I genuinely enjoy learning new technologies and reading insightful articles, and this curiosity is always a source of inspiration that I feel is actively utilized in my work.
Additionally, I often have casual discussions with colleagues outside of business hours. These are not necessarily about specific work tasks but are opportunities to exchange ideas, challenge each other’s thinking, and grow together. This collaborative culture supports both my professional growth and personal fulfillment.
AI Utilization
One of the important ongoing initiatives is the utilization of AI for operational efficiency. Specifically, we are advancing a project to automate repetitive and time-consuming tasks, such as AI training job log analysis and fault triage, using LLMs. We are currently developing a prototype, which is designed to identify potential root causes and suggest next steps using the model’s output. This will significantly streamline tasks that engineers previously debugged manually.
Although it is still under development, the initial results are very promising. We are confident that this initiative will not only contribute to a significant improvement in operational productivity and a reduction in downtime but will also serve as a powerful use case to promote broader AI adoption within Rakuten. We are driving innovation by actively applying AI in not only to product development but also to daily operational processes, thereby enhancing the productivity of the entire team.
*Semantic search model
A technology that enables search engines to go beyond simple keyword matching and deeply understand the searcher’s intent and the nuance/context of their query.
*The content and affiliation presented in this article are accurate as of the time of the interview.
For more information about available positions click here.
