【JAPAN AI】Agent Harness Engineer / English

900万円〜1,600万円

東京都

As an Agent Harness Engineer at JAPAN AI, you will design and implement the agent control and execution infrastructure, leveraging AI/ML knowledge. Key responsibilities include:

1. Agent Harness Design & Implementation:
- Design and implement the agent execution engine (Graph Runtime / State Machine)
- Develop the Agent SDK for in-house engineers
- Implement session management, checkpoint, and recovery mechanisms
- Build the guardrail/policy execution engine to control agent behavior

2. AI/ML System Integration:
- Implement model routing across LLM providers
- Design context management and memory infrastructure
- Optimize inference pipelines for latency and cost efficiency
- Integrate latest research findings into production infrastructure

3. Orchestration & Performance:
- Develop workflow orchestration and queuing systems
- Optimize cost and performance through autoscaling, caching, and batch processing
- Manage inference request routing and load balancing

4. Reliability & Operations:
- Maintain platform uptime of ?99.9%
- Conduct incident response and post-mortems
- Design data access and permission management infrastructure

The role focuses on building the core execution infrastructure for AI agents, serving as the critical control layer that transforms agents from demo prototypes to trusted production systems.

JAPAN AI株式会社

東京都新宿区西新宿6-8-1

正社員

健康保険 雇用保険 厚生年金 労災保険

完全週休二日制 定休日：土・日・祝日 休暇：年次有給休暇、夏季休暇（3日）、年末年始休暇（12月31日?1月3日）、慶弔休暇

2026/04/02