Associate Research Scientist

Job Type: Location Type: Remote: open to residents of any country

 

Job Description:

Rakuten Group, Inc. (TSE: 4755) is a global technology leader in services that empower individuals, communities, businesses and society. Founded in Tokyo in 1997 as an online marketplace, Rakuten has expanded to offer services in e-commerce, fintech, digital content and communications to 2 billion members around the world. The Rakuten Group has around 30,000 employees, and operations in 30 countries and regions. For more information visit https://global.rakuten.com/corp/.

Job Summary:

Rakuten Institute of Technology (RIT) is the corporate R&D department of the Rakuten group and is responsible for leading strategic technological progress. RIT’s targets vary from basic technologies to innovative new services. RIT consists of highly motivated research scientists located in Tokyo, Boston, Paris, Singapore, San Mateo and Bengaluru who are driven by strong curiosity.

We are seeking an innovative and dedicated research scientist to join RIT. As an Associate Research Scientist, you will play a pivotal role in shaping the future of AI within Rakuten’s diverse ecosystem. Our team focuses on enhancing in-house Large Language Models (LLMs) and seamlessly integrating them into Rakuten’s services, impacting millions of users worldwide. We are a collaborative and innovative group, working on cutting-edge research and fostering a culture of knowledge sharing. The ideal candidate will have a passion for producing groundbreaking research and collaborating with teams across Rakuten to deliver impactful AI solutions.

We welcome exceptional researchers and engineers from other fields who may not have direct experience in generative AI to apply. This role provides an opportunity to gain the critical skills and knowledge needed to transition into the AI & ML space. We are committed to supporting career changers who bring unique perspectives and expertise, and we offer mentorship and resources to facilitate your growth in the AI and ML domains.

Key Responsibilities:

• Develop and train LLMs, including data processing, pre-training, and fine-tuning. This also includes designing and running experiments, writing code, and evaluating results.
• Implement and verify latest technologies in Deep Learning, LLMs and AI agent fields, and reflect the implementation into in-house machine learning libraries.
• Work collaboratively with globally distributed people in a range of roles, including researchers, engineers, product managers, designers, and other key product stakeholders to accomplish complex tasks that deliver value to our business.
• Present findings and insights in team meetings and contributing to project documentation.
• Stay up to date on research and product advances in AI.

Qualifications:

• Proven experience in a research-focused role within industry, academia, or government institutions.
• Experience participating in machine learning projects, having completed the entire cycle of model training, experimental design, and evaluation.
• Basic knowledge and understanding of natural language processing, deep learning and generative AI, including LLMs, RAG, and AI agents, and a passion for staying at the forefront of these technologies.
• High-level coding skills in Python, with a strong understanding of software engineering principles.
• Experience in pre-training/fine-tuning existing LLMs (> 1B) using PyTorch.
• Strong communication skills to understand and effectively convey technical information.
• A strong emphasis on teamwork and a proactive attitude towards learning.
• Ability to speak, write, and communicate in Japanese is a plus
• Experience presenting as first author at top-tier peer-reviewed conferences.
• Familiarity with efficient training and inference techniques such as LoRA, pruning, knowledge distillation, and mixed precision training.
• Experience with distributed training frameworks such as DeepSpeed and FSDP.
• Experience with multilingual and multimodal modeling.
• Experience in developing software to process a large-scale dataset using a distributed computing framework (e.g., Hadoop)

Minimum Requirements:

• Masters in Computer Science, Machine Learning or related field with 2+ years of experience in relevant industry experience.
• PhD in in Computer Science, Machine Learning or related field preferred

Five Principles for Success
Our worldwide practices describe specific behaviors that make Rakuten unique and united across the world. We expect Rakuten employees to model these 5 Shugi Principles of Success.

Always improve, Always Advance – Only be satisfied with complete success – Kaizen
Passionately Professional – Take an uncompromising approach to your work and be determined to be the best
Hypothesize – Practice – Validate – Shikumika – Use the Rakuten Cycle to succeed in unknown territory
Maximize Customer Satisfaction – The greatest satisfaction for our teams is seeing their customers smile
Speed!! Speed!! Speed!! – Always be conscious of time – take charge, set clear goals, and engage your team

Rakuten provides equal employment opportunities to all employees and applicants for employment and prohibits discrimination and harassment of any type. Rakuten considers applicants for employment without regard to race, color, religion, age, sex, national origin, disability status, genetic information, protected veteran status, sexual orientation, gender, gender identity or expression, or any other characteristic protected by federal, state, provincial or local laws.

At the time of posting, Rakuten expects the base compensation for this role to be within the range shown below. Individual compensation will vary based on job-related factors, including the skills, qualifications, and experience of the successful candidate as well as business need and geographic location. The successful applicant for this role will be eligible for discretionary bonus, health, vision, dental insurance, 401k matching, PTO, Volunteer Time Off (VTO), and other employee benefits as the company implements.

USD $88,357.50 – $150,040.00 annually