Research Scientist (Generative & Agentic AI) - Tokyo,Japan.
At a glance
Mid-level AI Research role at Appier. Tokyo.
Pay not stated AI Research salary
aimlroles summary, based on the employer's posting.
What you'll do
- Build LLM- and VLM-powered agents that reason, plan, use tools, retain memory and collaborate with other agents
- Apply post-training methods such as SFT, RLHF and preference optimization to strengthen model quality, alignment and reliability
- Make foundation models faster and more scalable during training, inference and test-time computation
- Test models and agents against real-world tasks, then work with other teams to bring research into production
- Spot promising research directions and present your findings at major AI and machine-learning venues
What you bring
- A master's or PhD in computer science, electrical engineering, mathematics or a related subject, plus AI/ML research experience
- Expertise in foundation models, with depth in at least one area such as LLMs, multimodal models, reinforcement learning or agent systems
- Practical LLM-building experience, including fine-tuning, RAG, agent frameworks or product prototypes
- Proficiency in Python and PyTorch, with the ability to build, train and improve models
Who this fits
This role suits a researcher who can connect AI/ML research with deployed applications and collaborate with scientists and engineers. The job is based in Tokyo, Japan. Visa sponsorship is available for international applicants.
From the employer
About Appier
Appier is an AI-native Agentic AI as a Service (AaaS) company that uses artificial intelligence (AI) to power business decision-making. Founded in 2012 with a vision of democratizing AI, Appier’s mission is turning AI into ROI by making software intelligent. Appier now has 17 offices across APAC, Europe and U.S., and is listed on the Tokyo Stock Exchange (Ticker number: 4180). Visit www.appier.com for more information.
About the Role
As a Research Scientist, you will work at the frontier of generative and agentic AI: advancing Large Language Models (LLMs), Vision-Language Models (VLMs), and AI agents that reason, plan, and use tools to solve real-world problems. Your research will span post-training (SFT, RLHF/RLVR), reasoning and test-time scaling, multimodal intelligence, and autonomous agentic systems. You will shape Appier's core AI capabilities, publish at top AI/ML conferences, and collaborate with scientists and engineers to turn frontier research into product impact.
Responsibilities
- Research and build agentic AI systems: reasoning, planning, tool use, memory, and multi-agent collaboration, powered by LLMs and VLMs.
- Advance post-training techniques (SFT, RLHF, RL with verifiable rewards, preference optimization) to improve model capability, alignment, and reliability.
- Improve the performance, efficiency, and scalability of foundation models across training, inference, and test-time compute.
- Design rigorous evaluations and benchmarks for models and agents in real-world scenarios.
- Collaborate with cross-functional teams to ship research into production applications.
- Track frontier research, propose new directions, and publish key findings at leading AI/ML venues.
About You
Minimum Qualifications
- Master's degree or Ph.D. in Computer Science, Electrical Engineering, Mathematics, or a related field, with research experience in AI/ML.
- Deep understanding of modern foundation models, with expertise in at least one of: LLMs, VLMs/multimodal models, RL, or agentic systems.
- Hands-on experience building with LLMs: fine-tuning, RAG, agent frameworks (e.g., tool use, function calling), or product prototyping. Fluency with AI-assisted coding workflows is a plus.
- Proficient in Python and PyTorch; able to build, train, and optimize models effectively.
- Strong ability to analyze model behavior, diagnose bottlenecks, and improve training and inference pipelines.
- Clear communication skills and a team-first attitude in a fast-paced, collaborative environment.
Preferred Qualifications
- Publications in top AI/ML conferences (e.g., NeurIPS, ICML, ICLR, CVPR, ACL, EMNLP).
- Experience with large-scale distributed training or LLM post-training pipelines.
- Contributions to open-source projects (e.g., agent frameworks, model or benchmark releases).
- Passion for pushing the frontier of generative and agentic AI and bridging research with impactful real-world applications.
Open to overseas candidates/Visa Support
This position is based in Tokyo, Japan. For international candidates, Appier's Japan office provides visa sponsorship to ensure a smooth transition to Japan.