About the role
You will build, operate, and improve shared capabilities for developing and running LLM-powered applications, agents, and MCP integrations. You will design platform services, run Kubernetes infrastructure, support production use cases, improve developer experience, and own projects from architecture through operation.
Responsibilities
- Develop tooling and infrastructure for building, testing, deploying, and operating GenAI systems
- Design and operate services for model access, tool integration, orchestration, evaluation, observability, and runtime infrastructure
- Build reliable and scalable Kubernetes infrastructure
- Turn common engineering requirements into reusable platform capabilities
- Support complex GenAI use cases from experimentation through production
- Own architecture, implementation, deployment, and operation of platform projects
- Improve platform reliability, scalability, and standardization
- Contribute to technical designs, code reviews, and engineering standards
Requirements
- 4+ years of experience in platform engineering, DevOps, data platform engineering, MLOps, backend engineering, or ML engineering
- Linux system administration experience in cloud and on-premises environments
- Experience building or operating production LLM-powered applications, agents, or GenAI platforms
- Kubernetes, Helm, Argo CD, and GitOps experience
- Observability experience with Prometheus, Grafana, Kibana, or similar tools
- Python, backend service, and API proficiency
- Knowledge of LLM application architectures including tool calling, RAG, agent workflows, model gateways, and evaluation
- Experience diagnosing reliability, latency, scalability, cost, and deployment issues in GenAI systems
- Knowledge of security, access control, data privacy, and governance
Requirements
- 4+ years of experience in platform engineering, DevOps, data platform engineering, MLOps, backend engineering, or ML engineering
- Linux system administration experience in cloud and on-premises environments
- Experience building or operating production LLM-powered applications, agents, or GenAI platforms
- Kubernetes, Helm, Argo CD, and GitOps experience