MLOps

Position: MLOps Company: "Freedom Telecom Operations" LLP City: Astana Employment: hybrid Salary: 1,500,000 - 1,800,000 KZT

Freedom Telecom is a modern telecommunications company developing broadband internet access to every home and open Wi-Fi access in major cities in the Republic of Kazakhstan. We work in close cooperation with global equipment and software manufacturers. Job Description: Deployment of trained ML models into a production environment. Packaging models into Docker containers and configuring inference servers (Triton Inference Server, TorchServe, BentoML, or similar solutions). Ensuring model versioning and deploy process reproducibility. Preparation, setup, and administration of GPU servers, including installation and updating of NVIDIA Driver, CUDA, and cuDNN. Configuration of GPU nodes in a Kubernetes cluster. Monitoring GPU performance, diagnosing, and troubleshooting hardware issues. Building and maintaining CI/CD pipelines for automated testing, Docker image building, and ML service deployment. Setup and maintenance of monitoring and alerting infrastructure (Prometheus, Grafana, and similar tools). Monitoring ML service performance, tracking latency, availability, and computational resource usage. Managing GPU cluster resources, load balancing, and optimizing hardware utilization. Collaboration with Data Science, ML Engineering, DevOps, and Development teams during the implementation of machine learning models. Preparation of technical documentation and participation in the development of internal MLOps standards. Higher education in Information Technology, Computer Science, Applied Mathematics, or related fields. 2-3 years of experience in MLOps, DevOps, or ML Infrastructure. Hands-on experience with Docker, Kubernetes, and Helm. Understanding of containerization, orchestration, and resource management principles. Experience with ML model deployment platforms: MLflow, Triton Inference Server, BentoML, TorchServe, or similar solutions. Proficiency in Python for process automation, developing utility scripts, and working with APIs. Bash/Shell scripting skills. Experience in administering GPU servers, configuring NVIDIA Driver, CUDA, and cuDNN. Experience with Git and building CI/CD processes (GitHub Actions, GitLab CI, Argo Workflows, or similar tools). Experience setting up monitoring and alerting systems (Prometheus, Grafana). Understanding of the ML model lifecycle and MLOps principles.

Will be an advantage: Experience with Kubernetes GPU Operator. Experience operating high-load AI/ML services. Knowledge of distributed model training technologies. Experience with cloud platforms (AWS, Azure, GCP). Experience using Terraform or Ansible. Proficiency in Linux at the system administration level. English language proficiency for reading technical documentation.

Schedule: 5/2, from 9:00 to 18:00; Hibrid work format; Casual dress code; Privileges from the Freedom Holding ecosystem; Extra days off, in addition to vacation and sick leave; Material assistance for memorable events; Most innovative, ambitious projects and tasks; Career growth in a team of friendly professionals.

Contacts: Telegram recruiter contact Whatsapp