AI Infrastructure Engineer III

MOZN is a leading Enterprise AI company enabling organizations to make informed decisions in two critical domains: Financial Crime Prevention and Enterprise Knowledge Intelligence.

We’re a diverse, collaborative team of innovators united by a shared purpose: to build AI that delivers tangible business value, builds trust, and empowers people and organizations with augmented intelligence. Our culture is built on the relentless pursuit of excellence and meaningful impact.

If you’re passionate about working alongside exceptional talent on world-class AI, and you want the autonomy and runway to do the best work of your career, join us in shaping the future of intelligent enterprises.

We are looking for a highly motivated AI Infrastructure Engineer III to join our Cloud Engineering team. The ideal candidate is passionate about building scalable AI infrastructure that enables machine learning engineers and data scientists to efficiently develop, train, deploy, and operate AI models.

This role focuses on designing, operating, automating, and continuously improving AI platform capabilities including Kubeflow, MLflow, GPU infrastructure, distributed training platforms, model serving infrastructure, and MLOps tooling across cloud-native and hybrid environments.

The ideal candidate possesses deep expertise in AI infrastructure and GPU platforms. Kubernetes and cloud platform experience are essential enablers, but the primary focus is building highly optimized infrastructure for AI and machine learning workloads.

• Design, deploy, and operate enterprise AI/ML platforms. 

• Build self-service platforms for Data Scientists and ML Engineers. 

• Deploy and operate Kubeflow, MLflow, KServe, Ray, or similar AI platforms. 

• Design infrastructure supporting model training, experimentation, feature engineering, and inference. 

• Build highly available and scalable model serving infrastructure. 

• Design and operate GPU clusters for large-scale AI workloads. 

• Optimize GPU scheduling, utilization, sharing, autoscaling, and resource allocation. 

• Deploy and manage NVIDIA GPU Operator and GPU-enabled Kubernetes environments. 

• Optimize distributed GPU training performance across multi-node clusters. 

• Troubleshoot AI infrastructure performance bottlenecks. 

• Build CI/CD pipelines for ML workloads. 

• Automate AI infrastructure provisioning using Infrastructure as Code. 

• Implement monitoring and observability for GPU utilization, model serving, training jobs, and inference latency. 

• Collaborate closely with Data Science teams to improve platform usability, performance, and reliability. 

• 4-6 years of experience in AI Infrastructure, MLOps, Platform Engineering, or Cloud Engineering. 

• Strong hands-on experience with Kubernetes. 

• Experience with Kubeflow, MLflow, or similar ML platform technologies. 

• Experience operating GPU infrastructure for AI workloads. 

• Strong understanding of NVIDIA GPU technologies, CUDA fundamentals, and GPU optimization. 

• Experience supporting distributed training workloads. 

• Experience with model serving platforms such as KServe, Triton Inference Server, Ray Serve, or similar. 

• Experience with AWS, GCP, OCI, or Azure AI platforms. 

• Experience automating infrastructure using Terraform, Helm, GitOps, or Ansible.

• Strong scripting or programming skills in Python, Bash, or Go. 

• Experience with Prometheus, Grafana, OpenTelemetry, ELK/OpenSearch, or equivalent observability platforms. 

• Experience with PyTorch, TensorFlow, Hugging Face, or JAX. 

• Experience with distributed training frameworks such as Ray, DeepSpeed, Horovod, or NCCL. 

• Experience with Vector Databases, LLM infrastructure, RAG architectures, or GenAI platforms. 

• Experience operating inference platforms for large language models. 

• Experience supporting AI research or Data Science teams in production environments. 

• Contributions to Cloud Native, Kubernetes, AI, or ML open-source communities.

• Cloud, Kubernetes, NVIDIA, or AI/ML certifications are a plus.