About
Krisp develops production products including speech APIs, real-time streaming services, and voice AI platforms.
Responsibilities
•Maintain the reliability of core production services end to end, targeting a 99.9% SLA
•Design infrastructure architectures for multi-region, multi-cloud, private cloud, and on-premise enterprise deployments
•Design and build failover mechanisms between regions and cloud providers
•Build monitoring, alerting, and observability tools to detect issues, support incident management, and retain historical monitoring data
•Improve the stability and latency of WebSocket-based real-time streaming services, and optimize network costs and latency across regions
•Deploy new ML models and product features to production with engineering teams
•Lead production incident response, root cause analysis, and post-incident improvements
•Design, implement, and maintain CI/CD pipelines
•Manage system patching and security, and integrate security tools into the infrastructure
•Drive DevOps best practices, standards, and documentation across the team
Requirements
•5+ years of hands-on DevOps experience, including ownership of highly available production systems
•Experience designing and operating systems with strict SLA requirements of 99.9% or higher
•Deep production experience with Kubernetes and containers, including migrations from other orchestration platforms; EKS preferred
•Strong experience with at least one major cloud provider; AWS preferred
•Experience with multi-region and multi-cloud architectures
•Hands-on experience with Terraform and/or Ansible
•Experience designing CI/CD pipelines with Jenkins and GitHub Actions
•Strong experience with Prometheus, Grafana, or similar monitoring and observability stacks, and incident management processes
•Solid networking fundamentals, including TCP/IP, DNS, load balancing, VPNs, firewalls, and CDN
•Strong Linux system administration skills
•Proficiency in bash, Python, or similar scripting languages for automation
•Strong proficiency in Claude Code and AI-assisted coding workflows
Nice to have
•Private cloud or on-premise enterprise deployments
•Real-time or low-latency streaming systems (WebSocket, WebRTC, gRPC)
•Production ML model deployment and serving (GPU workloads, inference platforms)
•Cloud network cost optimization
Conditions
No conditions specified in the posting.
