Agent Harness Engineer / English (RT), Jobs in Tokyo - New Job Opening

Design the Heart of the Enterprise Brain. Build the Next-Generation Execution Engine and Guardrails for Advanced AI Agents!【Residents of Japan】

■ Job Title

Agent Harness Engineer

■ Company Overview

We are a fast-growing, elite AI startup specializing in cutting-edge enterprise SaaS platforms powered by Generative AI. Our product team manages a sophisticated environment supporting hundreds of concurrent workflows entirely in-house. We pride ourselves on a highly creative, collaborative environment where engineering and AI research intersect, providing our teams with massive scale and the freedom to design the industry's most advanced AI agent control frameworks.

■ Your Role and Responsibilities 

• Design and implement the core execution engine (Graph Runtime / State Machine) based on a deep understanding of LLM and AI agent operating principles.
• Own AI-specific system architecture, managing model routing, context processing, and memory infrastructure (both long-term and working memory).
• Design and develop a robust Agent SDK to be utilized internally by our engineering organization.
• Build a secure guardrail and policy execution engine to guarantee safe, quick, and reliable agent operations.
• Partner with Research Engineers to continuously integrate state-of-the-art research outcomes into our production infrastructure.

■ Experience and Qualifications

• Bachelor's degree or equivalent practical experience in Computer Science, Software Engineering, Artificial Intelligence, Machine Learning, Mathematics, Physics, or related fields.
• 5+ years of practical experience as a Backend Engineer.
• Strong hands-on production product development experience using Python.
• Proven experience designing and implementing production systems that leverage LLMs and AI agents.
• Hands-on experience designing and coding distributed systems from scratch (not just operating them).
• Professional experience designing and implementing RESTful APIs and gRPC.
• Language skills: Fluent Japanese (ability to discuss product development without friction) OR Business-level English.

■ Tech Stack

• Languages: Python, Go (Backend / Infrastructure), TypeScript / React / Next.js (Frontend), NX
• Infrastructure & Messaging: GCP (Containers / K8s), Docker, Terraform, Kafka, Pub/Sub
• Monitoring & Tools: Prometheus, Grafana, OpenTelemetry, GitHub, Slack, Confluence, Linear, Notion
• AI Dev Support: Claude Code MAX Plan, Cursor, ChatGPT, Devin
• Workstation: Mac (Apple Silicon), dual monitor setup

■ Good Reasons to Join

• **Elite AI Support:** Stay at the absolute cutting edge with full company coverage for top-tier AI dev tools, including Claude Code MAX, Devin, Cursor, and ChatGPT.
• **Learning and Growth:** Generous book purchase assistance (¥30,000 half-yearly), language learning programs, and qualification acquisition coverage.
• **Flexible Hybrid Culture:** Work 3 days in the office and 2 days remotely, supported by flexible hours and negotiable core times.
• **Exceptional Benefits:** Includes a monthly refresh allowance (¥5,000 for gym, movies, etc.), dedicated housing support (up to ¥30,000/month), and standard development tool support.
• **Financial Upside:** Competitive salary structure with stock options available, alongside biannual performance evaluations and bonuses.

■ Work Location

Tokyo (Hybrid Work: 3 days office, 2 days remote)

■ Salary 

¥12,000,000 〜 ¥20,000,000 (Based on experience; bonuses and stock options available)