SRE / Infrastructure Engineer, ML Platform

About this role

This SRE and infrastructure role owns the infrastructure supporting the MLOps Platform for Digital Bank and group fintech products. The platform includes serving interfaces, authentication, security layers, data pipelines, ML development capabilities, and production model endpoints for the Financier shared credit evaluation model initiative. The role focuses on infrastructure dedicated to the Digital Bank and fintech domain while working alongside the SRE responsible for the platform-wide foundation.

Key work includes building infrastructure and managing it as code with Terraform. The engineer will design, build, operate, and improve AWS infrastructure for serving interfaces, data pipelines, and the ML development platform. Responsibilities also include designing financial-grade network security, including mutual TLS authentication, AWS PrivateLink, VPC Peering, private connectivity for Digital Bank integration, and expanded security monitoring.

The role will establish observability for logs, metrics, and traces using CloudWatch and related tools. It includes defining SLOs and SLAs, creating an on-call rotation, standardizing incident response, and supporting reliable operation under high traffic and high availability requirements. The engineer will also design the interface between the Databricks data platform operated by the DRE organization and the MLOps Platform, covering data processing and pipeline integration.

The position collaborates with the SRE responsible for group-wide MLOps standards, data platform teams, product platform teams, and other group companies. The team is a scrum unit comprising a Product Manager, Engineering Manager, Data Scientists, and ML Engineers and is currently validating prototypes. The infrastructure objective is to support the planned FY2027 Digital Bank launch from a reliability and security perspective.

Required experience includes approximately five or more years as an SRE or infrastructure engineer; hands-on AWS design, construction, and operations using services such as ECS, API Gateway, VPC, and IAM; Terraform experience covering infrastructure as code, module design, and CI/CD automation; and advanced networking and security architecture using VPC Peering, AWS PrivateLink, and TLS or mutual TLS. Candidates should also have experience with cloud observability, monitoring, alert design, performance log analysis, CI/CD, containers, GitHub Actions, and Docker.

Preferred experience includes infrastructure support for MLOps or data platforms using Amazon SageMaker, Databricks, or Airflow; defining SLOs and SLAs and standardizing incident response with tools such as Datadog and PagerDuty; financial-services or payments security and audit work aligned with the FISC Security Guidelines; and shared-platform work across multiple organizations. The stated technology stack includes AWS, ECS Fargate, API Gateway, VPC, PrivateLink, SageMaker, S3, Glue, CloudWatch, Databricks, Terraform, GitHub Actions, Docker, Python, Shell, SQL, Claude Code, Slack, and Notion. Datadog and PagerDuty have not yet been adopted, so observability and incident-management tool decisions will be part of the role.

This is a full-time position in Tokyo. Annual salary is JPY 7,008,000 to JPY 11,004,000, with monthly salary of JPY 584,000 to JPY 917,000. The salary includes fixed allowances for up to 45 hours of legal overtime, legal holiday work, and 40 hours of late-night work. A high-performance bonus may be paid based on semiannual evaluations. The probation period is three months. The working arrangement uses a discretionary labor system for professional work, with basic hours of 9:30 to 18:30 and a 60-minute break; employees may choose their working hours at their discretion, and the system may change to flextime. The hybrid policy requires at least two office days per week as standard. Saturdays, Sundays, Japanese national holidays, paid leave, summer and winter holidays, and year-end and New Year holidays are provided. Benefits include social insurance, housing and moving allowances, health checks, influenza vaccination, book-purchase support, a defined-contribution pension, an employee stock ownership plan, service discounts, work equipment support, a company library, recruitment-meal coverage, referral rewards, and partial support for domestic and international conferences. Business-level Japanese equivalent to JLPT N2 or above and basic business-level English equivalent to TOEIC 700 or above are required; interviews are conducted in Japanese. Only applicants residing in Japan are eligible to apply.