sre

The company develops an advanced platform for creating and managing AI agents. This solution supports deployment within customer infrastructure as an enterprise product, as well as a SaaS version. The platform integrates real-time voice and telephony, GPU and LLM inference, and streaming analytics, operating in both cloud and on-premise environments.

• Ensure service reliability by managing SLIs/SLOs, availability, and eliminating system bottlenecks;
• Set up monitoring, metrics, alerts, and dashboards;
• Build and maintain Grafana dashboards for internal teams and customers;
• Conduct load testing, analyze results, and provide scaling recommendations;
• Investigate incidents, participate in on-call rotations, and lead postmortems;
• Collaborate with developers to challenge technical decisions and find solutions;
• Develop and support Kubernetes-based infrastructure on GCP and AWS;
• Automate routine tasks and assist with CI/CD processes;
• Deliver and support the platform for customers, including on-prem deployments;
• Mentor colleagues and contribute to raising the engineering standards of the team.

• 5+ Years of experience in SRE/DevOps with responsibility for high-load production systems;
• Deep practical understanding of Docker and Kubernetes in production;
• Hands-on experience with metrics, alerts, Prometheus, Alertmanager, and Grafana;
• Experience with SLIs/SLOs, incident investigation, and postmortems;
• Experience with load testing and capacity planning;
• Proficiency in Python for automation and tooling;
• Experience with GCP and/or AWS, strong Linux skills, and solid networking knowledge;
• Knowledge of DevOps fundamentals including CI/CD, GitHub Actions, Terraform, and Ansible;
• Ownership mindset, analytical thinking, and proactive approach to outage prevention;
• Strong communication and mentoring skills;
• Nice to have: Experience using AI agents, real-time telephony (SIP, FreeSWITCH, RTP, WebRTC), GPU/ML serving (Triton, vLLM, RunPod, Nebius, Lambda, run:ai, DCGM), streaming data and analytics (Kafka, ClickHouse), IaC and GitOps (ArgoCD), logging (Loki/ELK), gRPC, working in isolated secure environments, and preparing systems for significant load growth.

• 21 Vacation days plus public holidays and 5 sick days;
• Private English lessons via Preply.