MLOps/AIOps SRE - Software Engineer(Enterprise Networking)
Description
工作說明
We are looking for an MLOps/AIOps-focused Site Reliability Engineer (SRE) to build and maintain reliable, automated, and scalable pipelines and infrastructure for our AI/ML systems supporting enterprise networking products. This role bridges DevOps, machine learning lifecycle management, and cloud backend reliability to ensure our AI-driven services operate efficiently and accurately.
What You’ll Do:
1. Build, maintain, and automate end-to-end ML pipelines including data ingestion, feature processing, model training, validation, deployment, and monitoring.
2. Implement model serving and scalable inference systems on Kubernetes or similar platforms.
3. Develop model performance monitoring, drift detection, and alerting mechanisms.
4. Collaborate closely with AI/ML engineers to optimize training workflows, model packaging, and reproducibility.
5. Implement AIOps workflows, integrating observability data, anomaly detection, ML-based insights, and automated remediation.
6. Automate management of ML artifacts, datasets, and feature stores.
需求條件
擅長工具：Python
【Essential Requirements】
1. Strong scripting skills (e.g., Python, Bash).
2. Hands-on experience with ML workflow orchestration, such as Kubeflow, MLflow, Airflow, or similar.
3. Proficiency with Git and modern CI/CD systems (e.g., Jenkins, GitHub Actions, GitLab CI).
4. Solid experience with Docker and Kubernetes in production environments.
5. Experience with monitoring/logging systems (Prometheus, Grafana, ELK).
6. Familiarity with Infrastructure-as-Code using Terraform or equivalent.
7. Understanding of ML model lifecycle, versioning, packaging, and deployment.
【Preferred Qualifications】
1. Deep knowledge of MLOps frameworks (e.g., MLflow, Kubeflow Pipelines, Feast, BentoML, Seldon).
2. Experience with Golang, especially for performance tools, operators, or controllers.
3. Experience implementing AIOps systems, such as anomaly detection from metrics/logs/traces, automated remediation, AI-driven incident prediction.
4. Experience integrating ML with network analytics, telemetry pipelines, and edge/cloud inference.
5. Familiarity with Cilium, Falco, or other cloud-native security and observability tools.
6. Experience with GPU-accelerated workloads, distributed training frameworks, or data engineering pipelines.
Skills
- AI
- Airflow
- Analytics
- Anomaly Detection
- Bash
- BentoML
- CI/CD
- Cloud
- Cloud Native
- Data Engineering
- Data Ingestion
- DevOps
- Distributed Training
- Docker
- ELK
- Git
- GitHub
- GitHub Actions
- GitLab
- Go
- GPU
- Grafana
- Infrastructure as Code
- Jenkins
- Kubeflow
- Kubernetes
- Machine Learning
- MLflow
- MLOps
- Networking
- Observability
- Prometheus
- Python
- SRE
- Terraform
- Workflow Orchestration