MLOps Engineer

We are looking for a MLOps Engineer to join a fast-growing, AI-driven technology company working on large-scale, production-grade machine learning systems, including LLMs, speech, and streaming models.

Requirements

• Deploy and host machine learning models
• Serve models in production environments
• Optimize throughput, GPU utilization, and resource efficiency
• Manage and operate multiple models concurrently in production
• Benchmark and profile model performance to ensure continuous optimization
• Build and manage containerized workloads using Docker and Docker Compose
• Deploy and operate production systems on Kubernetes
• Use Helm to deploy services, metrics, and supporting infrastructure
• Implement continuous deployment workflows using Flux CD
• Design auto-scaling and serverless serving solutions using KEDA, KServe, and Knative
• Ensure network security, traffic routing, and load balancing within Kubernetes environments
• Work extensively with AWS services
• Manage infrastructure and Kubernetes clusters using Terraform
• Configure and maintain GPU-enabled nodes
• Handle certificate management and system-level security
• Operate and troubleshoot Linux-based systems
• Design and manage asynchronous data pipelines using Kafka
• Work with Redis and SQL-based data stores for caching and persistence
• Implement monitoring and observability using Datadog
• Maintain and extend internal services and pipelines using Python
• Build and manage CI/CD pipelines with GitHub Actions
• Use workflow orchestration tools such as Flyte for model fine-tuning and ML workflows