Principal Engineer (AI Agent Platform)

About the project

The AI Agent Platform is a strategic foundational platform to enable efficient development and operation of multiple AI agents across Money Forward. It acts as a common engine where many AI agents can be developed, managed, and operated in a unified way, delivering consistent, high-quality experiences to our users and supporting the goals of MF AI Vision.

As a Principal Engineer, you will be a core member of the AI Agent Platform development team, responsible for designing and implementing large-scale backend and platform capabilities (agent lifecycle management, configuration & data pipelines, integrations with internal/external systems, observability, etc.) and driving technical excellence across teams.

Responsibilities

• Design, build, and evolve core services of the AI Agent Platform (APIs, backend services, internal tools) with a focus on scalability, reliability, and maintainability for multiple AI agents and products.
• Lead end-to-end system design for new platform capabilities (e.g., rulebook / knowledge management, vector DB integration, authentication & authorization, observability), producing clear architecture documents and reviewable RFCs.
• Collaborate closely with PdM, architects, AI engineers, QA, SRE, and product teams to translate business and product requirements into robust technical designs and implementation plans.
• Drive technical roadmapping and versioning of the platform (e.g., defining scope and milestones per platform version, aligning them with product release plans).
• Ensure high-quality API and data contracts between the platform and AI agents / products, including clear ownership boundaries and migration paths.
• Provide technical leadership and mentoring for other engineers (backend, frontend, AI), raising the bar in code quality, design, and operational excellence.
• Work with QA and SRE to design and implement testing strategies, capacity planning, and incident response for the platform’s services and environments.
• Contribute to and sometimes lead cross-team technical discussions and RFC reviews around security, data policies, guardrails, and cross-project integrations (e.g., Mepar, Anyflow, MFCon, etc.).
• Support a healthy engineering culture by promoting documentation, knowledge sharing, and continuous improvement in ways of working (sprint planning, backlog management, retrospectives, etc.).

Required qualifications

We are looking for a candidate with a background similar to our current Principal Software Engineer in the AI department:

• 10+ years of professional software engineering experience, including several years in senior/principal roles with significant system ownership (tech lead or equivalent).
• Strong experience in full-stack web or backend development and large-scale system design, including microservices, distributed systems, and high-traffic APIs.
• Proficiency in at least one of the following major languages used in our AI teams:
  • TypeScript / JavaScript (Node.js)
  • Java
  • Python
  • C++
• Hands-on experience with relational databases and data pipelines (designing schemas, data flows, and batch/stream processing for product or analytics use cases).
• Proven ability to lead complex technical initiatives end-to-end: requirement clarification, architecture proposal, implementation planning, risk management, and leading others through execution.
• Experience working with cross-functional and offshore teams, including aligning with stakeholders across multiple departments and locations.
• Solid understanding of software engineering best practices: code review, testing strategies, CI/CD, observability, and secure coding.
• Business-level English communication (written & spoken) to collaborate with Japanese and global stakeholders; willingness to work in a bilingual environment.

Additional qualifications

• An active implementer capable of translating requirements into AI-driven workflows, rapidly executing everything from prototyping to initial implementation and deployment.
• Experience building or integrating with LLM / AI agent platforms, vector databases, or RAG-style systems.
• Knowledge of MCP (Model Context Protocol), AI SDKs, or similar agent integration frameworks.
• Background working on multi-tenant SaaS platforms, with clear separation of tenant data, authN/authZ, and RBAC models.
• Experience with Azure or other major public cloud providers, including infrastructure-as-code, networking, and security guardrails.
• Prior experience in FinTech / B2B SaaS domains, especially ERP/accounting or back-office systems, is a plus.
• Familiarity with engineering processes such as RFC-driven development, version definitions (VD), and structured project management (Jira, WBS, etc.).