5 months ago
Tokyo, JapanSenior / Staff+
Responsibilities
- Architect, implement, and scale Spark batch and Flink streaming pipelines for ML training and evaluation.
- Build and operate job execution frameworks for model training, inference, and post-processing.
- Develop internal API servers and tools to orchestrate ML jobs on Kubernetes using Argo Workflows, Helm, and Terraform.
- Design and monitor ClickHouse and PostgreSQL data infrastructure.
- Ensure platform availability and observability with Prometheus and Grafana.
- Collaborate with data scientists, product managers, and engineers to deliver reliable ML platform capabilities.
- Promote LLM-based tools such as GitHub Copilot and ChatGPT for development, documentation, and debugging.
- Mentor junior engineers and help evolve engineering culture and standards.
Requirements
- Bachelor’s degree in Computer Science, Engineering, or a related field; a Master’s degree is preferred.
- At least four years of hands-on experience in data systems, machine learning infrastructure, or platform engineering.
- Strong Python and/or Java coding proficiency and experience building large-scale production systems.
- Practical experience with Spark, Flink, Kubernetes/GKE, Terraform, and Helm.
- Experience managing high-throughput data infrastructure with ClickHouse, PostgreSQL, or similar systems.
- Deep understanding of production machine learning pipelines and distributed job execution.
- Experience applying LLM-based tools such as Claude Code and Codex to improve engineering productivity.
- Strong ownership, architectural thinking, and ability to lead cross-functional platform projects.
Tech Stack
Categories
About Appier
Appier builds AI-driven advertising and marketing software for enterprises, offering programmatic ad bidding, customer data, and predictive tools delivered as SaaS. Founded in 2012 and headquartered in Taipei, it is a public company listed on the Tokyo Stock Exchange (ticker: 4180). Its Ad Cloud processes millions of bid requests per second across APAC, Europe, and the U.S., and is used by in-house and agency marketers.