Senior Software Engineer, Agentic AI Platform - Mercari

JD in Japanese follows. 英文の後に和文JDをご覧いただけます。

Senior Software Engineer, Agentic AI Platform - Mercari

• Employment Status: Full-time
• Work Hours: Full Flextime (no core time)
• Office: Roppongi

For more details, see the Overview of Our Positions section on our Careers site.

About Mercari

Circulate all forms of value to unleash the potential in all people

"What can I do to help society thrive with the finite resources we have?" The Mercari marketplace app was born in 2013 out of this thought by our founder Shintaro Yamada as he traveled the world. We believe that by circulating all forms of value, not just physical things and money, we can create opportunities for anyone to realize their dreams and contribute to society and the people around them. Mercari aims to use technology to connect people all over the world and create a world where anyone can unleash their potential. For more information about Mercari Group’s mission, see Mercari’s Culture Doc

Organization/Team Mission

Mercari Engineering Principles

Mercari Engineering Principles are a shared understanding that serves as the foundation of engineering beliefs and behavior at Mercari. The Engineering Principles are designed to complement the organizational identity (Mercari’s mission, values, and culture) from an engineering viewpoint.

These principles ultimately help us achieve Mercari’s mission by defining the ideal state we seek to realize in the long term.

• Passion For The Product
• Grow Together
• Solve Through Mechanisms
• Collaborate Openly

For more details, please see the following link:

• Engineering Culture

The AI Platform team drives Mercari Group's transformation into an "AI-Native company" from the perspective of technical foundations. We design, develop, and operate shared AI infrastructure used across the company, including an LLM Gateway, MCP Gateway, and Agent Sandbox, and we also build the AI Agents that work alongside development and operations on the ground.

What we build is used in two ways. First, as services that every role—including engineers, designers, and PMs—uses directly in their daily work. Second, as shared building blocks that other engineering teams inside Mercari use to construct AI workflows tailored to their own domains. For the former, we create an environment where every role can demonstrate AI as a baseline skill, through developing AI Agents and running PoCs for AI SaaS. For the latter, we provide foundations with enough generality for other teams to extend on their own, and enough reliability for them to trust with real work.

Where a product team delivers customer value directly, we make other teams able to deliver that same value themselves. That is why we measure ourselves not by the number of things we have built, but by what the people using them became able to do. In this role, you will work as an enabler for AI-native development across Mercari Group while owning the shared AI platform it runs on: you build the mechanisms that let other teams work AI-natively, and you operate the production services those mechanisms depend on.

See here for more information about our mission and values.

Work Responsibilities

You will be responsible for designing, developing, and operating the technical foundations that enable AI adoption across Mercari Group.

• Design, develop, and operate the AI Platform: Design, implement, and operate shared AI infrastructure such as the LLM Gateway, MCP Gateway, and Agent Sandbox. These are production services the whole company depends on, so you own their operation—availability, security, and cost optimization—alongside the enablement work. Decide what to offer as company-wide infrastructure by weighing the range of teams affected, the impact on development as a whole, and system-wide risk.
• Build AI Agents and an execution platform: Develop AI Agents that work alongside development and operations, and build an execution platform that shifts the developer experience toward AI-native. Design on the premise that both humans and AI Agents are users: clear interfaces, documentation and metadata prepared in a form AI can interpret, guardrails implemented as code, and mechanisms that continuously evaluate quality.
• Own adoption and outcomes: Continuously measure whether the foundations you shipped are actually being used and how the development process changed for the people using them, and when adoption does not spread, identify the cause and see the improvement through.
• Promote company-wide AI adoption: Partner with business and product teams to embed AI-first workflows across the organization through AI SaaS PoCs and community operations. Operate company-wide shared repositories as innersource, responding continuously to pull requests from other teams.

Development Environment

• Infrastructure: Google Cloud Platform
• Languages: Go, TypeScript, Python, etc.
• Code management: GitHub
• Configuration management: Terraform
• Analytics: BigQuery / Looker
• Knowledge: Notion, Confluence, Linear
• AI development tools: Claude Code, Codex, Cursor, etc.
• AI/LLM development: Foundation model APIs, LiteLLM, etc.

Note: Technologies may differ depending on the area of responsibility.

Unique Challenges

• Enable every role at Mercari to work AI-natively: Enabling thousands of engineers, designers, and PMs to make AI a baseline skill, by building the shared foundations they build on top of and the agents they use in their daily work. Your impact is measured by what those teams became able to do without you in the room, not by the number of tools you shipped. You will work at a scale where you redefine how development is done across Mercari Group, not merely provide tools.
• Implement and operate a rapidly evolving AI tech stack at Mercari scale: Tackle advanced technical challenges by operating emerging stacks such as LiteLLM, MCP, and AI Agents safely and scalably in an organization of thousands—an initiative that is globally pioneering.
• Design foundations for the agent era: Design infrastructure, architecture, and operating models from scratch for a future where humans and AI agents collaborate, and build knowledge ahead of the industry. A rough edge that a person working by hand would simply overlook becomes a large difference in failure rate once AI Agents repeat the same steps at volume. Designing with that difference built in is territory where no established answer exists yet, inside or outside the company.
• Evolve the platform through collaboration with diverse stakeholders: Work with various internal organizations (Platform, Product, Security, etc.) to lead technology selection, standardization, and adoption.
• Develop and roll out services for internal developers: You can lead everything end-to-end—from planning internal services for Mercari engineers to supporting adoption. Rather than being only a "supporter," you are expected to proactively identify and solve problems as the person who best understands internal developers' pain points. Instead of answering requests one by one, or writing guidelines and calling it done, you own the whole loop: finding the problem that keeps recurring, turning it into a reusable mechanism, driving adoption, measuring what changed, and improving on it. That also means judging what is worth turning into a mechanism at all—weighing how many teams it affects, the critical path, and system-wide risk—and declining, with your reasoning, requests that cost a lot and help only one team. This role is not a reactive support desk that measures itself on response time, and not a gatekeeper that uses standards as a means of control.
• Work as a user of the platform you build: Use your own foundations in your own daily development, and join the teams that use them to understand their problems firsthand. Rather than imposing standards and processes as constraints, you can aim for a state where they get chosen because they are the faster way to get the work done.

Qualifications

…