over 1 year ago
Taipei, TaiwanSenior
Responsibilities
- Architect and operate resilient ML job execution frameworks for training, inference, and post-processing.
- Develop API services and developer tooling to orchestrate ML workflows on Kubernetes using Argo Workflows, Helm, and Terraform.
- Build scalable batch pipelines with Apache Spark for large-scale ML training and evaluation.
- Design and maintain data infrastructure using Trino, Databricks, and modern database technologies.
- Implement monitoring and observability with Prometheus and Grafana.
- Develop tooling that accelerates ML experimentation and production workflows.
- Collaborate with ML scientists to turn research prototypes into reliable, scalable, user-facing AI products.
- Lead cloud infrastructure design and operations on GCP using managed services including GCE, GKE, Cloud Storage, Cloud Functions, Cloud Pub/Sub, Cloud SQL, and BigQuery.
- Define and implement automated CI/CD pipelines using Jenkins, GitHub Actions, or ArgoCD.
- Apply distributed computing and parallel programming to optimize resource utilization and system performance.
Requirements
- Bachelor’s degree in Computer Science, Engineering, or a related technical field; a master’s degree is preferred.
- At least 5 years of hands-on experience in ML platform engineering, MLOps, or data infrastructure deploying enterprise-grade ML systems at scale.
- Expert proficiency in Python, Java, or Go, with strong data structures and algorithm foundations.
- In-depth experience with AWS or GCP and cloud-native service management.
- Strong experience with Docker and Kubernetes cluster management, including provisioning, autoscaling, and deployment practices.
- Understanding of the full ML lifecycle, including training, prediction, evaluation, backtesting, and feedback loops.
- Familiarity with Git workflows and Linux development environments.
- Experience with MarTech or customer-centric products is preferred.
- Demonstrated architectural leadership and ownership of complex cross-team platform initiatives.
- Experience with deep learning and ML workflow platforms such as Kubeflow, MLflow, or AWS SageMaker is preferred.
- Experience with Apache Spark and workflow orchestration tools such as Apache Airflow, Argo Workflow, or Luigi is preferred.
- Experience with production ML applications, real-time inference architectures, batch processing trade-offs, and large-scale training and serving optimization is preferred.
- Experience building applications related to LLMs, multi-agent LLM systems, or NLP is preferred.
Tech Stack
Apache AirflowApache SparkAWSDatabricksDockerGitGitHub ActionsGoGoogle BigQueryGoogle Cloud PlatformGrafanaHelmJavaJenkinsKubernetesLinuxMLflowPrometheusPythonTerraform
Categories
Data EngineeringML Engineering
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.