Kaltura
Kaltura
Kaltura develops a video cloud platform for organizations that use video to communicate, teach, connect with audiences, and deliver digital experiences. Its Video Experience Cloud brings together tools for managing, publishing, streaming, and conferencing content, supporting use cases such as virtual events, internal communications, education, and entertainment. For job seekers, Kaltura offers the opportunity to work on products that shape how organizations create, distribute, and engage with video across channels.

DevOps Engineer, AI Factory (Hybrid, Israel) – Kaltura

Build and operate reliable, GPU-powered AI infrastructure for Kaltura’s global video platform. Manage multicloud environments spanning AWS, GCP, Kubernetes, and AI serving technologies.

Description

  • Join Kaltura’s global DevOps and Platform team to design and run distributed AI production systems at significant scale
  • Lead projects from initial design through implementation and production deployment
  • Design, build, and operate highly available distributed systems across AWS and GCP
  • Support Kubernetes, Docker, Helm, CI/CD, automation, data, caching, and messaging infrastructure
  • Unify varied technology stacks and architectures within Kaltura’s broader platform
  • Develop and operate infrastructure for AI workloads, including GPU inference and model-serving platforms
  • Improve system reliability through monitoring and observability practices
  • Investigate complex problems across application, infrastructure, networking, operating system, and data layers
  • Partner with Platform, Development, Architecture, AI/ML Research, and Product teams
  • Use contemporary AI coding tools to increase engineering speed and quality

Requirements

  • At least 2 years of hands-on DevOps, Platform, or SRE experience operating large-scale distributed production systems, with the ability to independently lead projects from conception through implementation
  • At least 2 years of experience using cloud inference services and NeoClouds
  • At least 2 years of hands-on experience developing and maintaining AI Factory systems with KubeFlow, Ray, or comparable technologies
  • At least 2 years of practical experience with GPU inference using technologies such as vLLM, SGLang, or NVIDIA Triton
  • Extensive multicloud experience across AWS, GCP, and NeoCloud providers such as RunPod, TogetherAI, and vast.ai, including Infrastructure as Code with Terraform or a comparable tool
  • Advanced hands-on knowledge of Kubernetes and Helm, including Kubernetes internals, production operations, Docker, and containerized systems
  • Strong command of networking and operating systems, including TCP/IP, DNS, load balancing, and Linux internals
  • Substantial experience with CI/CD and automation using GitHub Actions, plus monitoring and observability with metrics, tracing, logging, Prometheus, and Grafana
  • High proficiency with modern AI coding tools and a practical understanding of how to use them for meaningful productivity gains
  • Familiarity with AI evaluation and observability platforms such as Opik, OpenEval, or LangSmith is a plus

Benefits

  • Flexible hybrid working environment
  • Extended private health insurance, including mental health coverage
  • Programs supporting personal and professional development
  • Occasional cross-company long weekends

Related Jobs

Veradigm®

Digital Client Success Manager

Veradigm®
1,001 – 5,000 Employees
HealthcareHealthcare InsuranceSaaS

Manage a digitally supported portfolio of healthcare technology customers at Veradigm, using customer health signals and structured playbooks to guide engagement. Support adoption, retention, renewals, and growth through risk management and coordination with internal teams.

Open