Senior SRE: AI Platform Reliability & MLOps

Tiger Analytics Inc.; Washington (District of Columbia); Híbrido; USD 150.000 - 190.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Career growth opportunities

Descripción de la vacante

Tiger Analytics Inc. is seeking a Site Reliability Engineer (SRE) to join our Forward Engineering team. You will safeguard production AI platforms, ensuring resilience, scalability, and performance across data-driven workloads.

Hybrid role combining software engineering and systems architecture with a focus on MLOps to bridge model development and production reliability. Strong Kubernetes, GPU/TPU optimization, and automation skills are required.

Formación

• Proven experience as an SRE or similar role
• Strong knowledge of Kubernetes, Docker, and cloud infrastructure
• Experience with IaC, CI/CD, and GitOps practices
• Proficiency in Python; Go is a plus
• Familiarity with ML platforms (Vertex AI, Kubeflow) and ML pipelines

Responsabilidades

• Define/SLOs for AI/ML services and monitor reliability
• Manage error budgets balancing velocity and stability
• Architect auto-scaling for Kubernetes (GKE) and model training workloads
• Ensure high availability of Vertex AI endpoints and inference services
• Optimize resource use for GPUs/TPUs and large language models
• Improve CI/CD pipelines and IaC provisioning
• Develop self-healing scripts and automation to reduce toil
• Lead incident response and post-mortem analysis

Conocimientos

Kubernetes

Docker

Python

Go

IaC

CI/CD

GitOps

Observability

Incident response

Herramientas

Kubeflow

Vertex AI

MLflow

DVC

GKE

Prometheus

Grafana

Stackdriver

Descripción del empleo

Tiger Analytics Inc. is seeking a Site Reliability Engineer (SRE) to join our Forward Engineering team. You will safeguard production AI platforms, ensuring resilience, scalability, and performance across data-driven workloads.

Hybrid role combining software engineering and systems architecture with a focus on MLOps to bridge model development and production reliability. Strong Kubernetes, GPU/TPU optimization, and automation skills are required.