AI Agent

Company Overview And Recruitment Background Equmenopolis Inc. is on a mission to revolutionize human-AI coevolution. We are pioneering an AI-driven diagnostic assessment and learning assistance platform, unlocking human potential and transforming education and business sectors. Our flagship product, LANGX, is a next-generation AI-powered language learning system that analyzes and enhances English proficiency through dynamic conversations with AI agents. Initially adopted as part of Waseda University’s official English curriculum, its impact is now rapidly expanding across educational institutions nationwide in Japan. Recognized for its groundbreaking innovation, LANGX has secured prestigious global accolades, including the Best Paper Award at SIGDIAL 2024, the SXSW EDU Launch 2023 Finalist Selection, the University Startup Award 2024, and the JEITA Startup Award establishing Equmenopolis as a global leader in AI-powered assessment and learning technologies. As we scale, we are seeking exceptional talent to drive the evolution of our platform. Our R&D team is an elite force of top-tier specialists in multimodal conversational AI, natural language processing, linguistics, learning science, digital humans, data science, software engineering, UI/UX and project management. With experts from over 10 countries, we are an international powerhouse, harnessing diverse expertise to develop our innovative products. As AI evolution accelerates at an unprecedented pace, Equmenopolis is redefining the relationship between humans and AI, taking on the challenge of transforming society itself. For engineers, researchers, and designers,

we offer
an exhilarating environment that goes beyond mere technological development — an opportunity to shape the next society. Main Responsibilities In this role, you will serve as a core contributor to our conversational diagnostic AI agent platform "EQU AI," responsible for designing, building, and operating production-grade LLM and multi-agent systems. Working closely with research and product teams, you will play a critical role in delivering human-Agentic AI collaborative problem-solving solutions to society.
• Agentic AI System Design & Implementation: Design and implement autonomous agent workflows to solve complex multi-step tasks, including Planning, Tool Calling, Memory, and Multi-Agent Orchestration.
• Model Selection and Adaptation (Adaptation/Fine-tuning): Comparative evaluation and integration of LLM / VLM (Vision-Language Model). Select and integrate optimal foundation models based on rigorous tradeoff analysis of Quality, Latency, and Cost. Build and operate fine-tuning pipelines using state-of-the-art techniques such as LoRA, Instruction Tuning, and Preference Optimization to pursue domain-specific performance.
• Prompt Architecture Design and MLOps: Develop System Prompt and RAG strategies prioritizing Safety and Robustness. Build data pipelines leveraging Vector Store and implement Prompt Versioning to manage prompt and data changes, driving LLM MLOps forward.
• Evaluation & Experiment Infrastructure Development and Operations: Design and operate infrastructure for offline/online evaluation of AI agent outputs, A/B testing, and Human-in-the-Loop evaluation to ensure reliability.
• Core Function Agent Development: Design and implement autonomous agent flows for automatic generation of dialogue scenarios, diagnostic and feedback generation for learners, and quality assurance of content, supporting the foundation of the product. Create papers, patents, and internal technical documentation, and present at conferences and workshops as needed. Required Qualifications And Skills
• Production-Grade LLM Product

Experience:
Hands-on experience deploying and operating LLM-powered software for complex multi-step tasks beyond PoC level, reaching actual production environments.
• Practical Agent Engineering

Experience:
Experience designing, implementing, and operating agent-specific components such as Tool Calling, Planning, and Multi-Agent Coordination in production environments.
• Data-Driven Model Selection Capability: Ability to select optimal LLMs for specific use cases through evidence-based tradeoff analysis based on KPIs such as Latency, Quality, and Cost.
• LLM Adaptation Technology

Experience:
Practical experience with LLM fine-tuning and adapter techniques such as LoRA and Instruction Tuning.
• Strong Software Engineering Fundamentals: Development skills in Python or TypeScript with emphasis on unit testing, CI/CD, specification, and code readability.
• Communication skills in Japanese (JLPT N1 or equivalent) Desirable Qualifications
• LLM MLOps / Infrastructure

Experience:
Experience building and operating LLM-specific MLOps components such as Vector Store, Feature Store, and Prompt/Data Versioning, as well as experience building and operating cloud infrastructure (GCP/AWS), containerization (Kubernetes/Docker), and observability (Monitoring/Tracing/Logging).
• Multimodal / Voice / Dialogue System Expertise: Deep knowledge of voice processing technologies such as Speech-to-Text, Text-to-Speech, Voice Activity Detection (VAD), and latency optimization, or advanced NLP/dialogue system development experience including RAG, Embedding model selection, and dialogue quality evaluation.
• Reliability & Evaluation

Experience:
Experience designing and operating evaluation systems for dialogue systems, including automated evaluation and Human-in-the-Loop setups, as well as experience with safety, red-teaming, and policy/guideline enforcement in LLM systems. Employment Type Full-time employee Work Location 3F, Green Computing Systems Research Center, 27 Waseda, Shinjuku-ku, Tokyo Nearest station: 1-minute walk from "Waseda Station" on Tokyo Metro Tozai Line Work Style Hybrid Desired Start Date Anytime Visa Sponsorship Visa sponsorship available Company Details Headquarters Entrepreneurship Center, 1-22-3 Nishi-Waseda, Shinjuku-ku, Tokyo Research Lab 3F, Green Computing Systems Research Center, 27 Waseda, Shinjuku-ku, Tokyo Established May 2, 2022 CEO Yoichi Matsuyama Main Business Development of conversational AI agent platforms and their applications URL https://www.equ.ai/en/

How to apply
Please enter "AI Agent /LLM Engineer" in the job title field and apply through the application form below.