The Red Hat AI Stack Engineer / Architect is responsible for designing, implementing, and operating AI and GenAI solutions on Red Hats AI platform, including Red Hat OpenShift AI (RHOAI) and Red Hat AI Enterprise, across hybrid and multi-cloud environments.

They work closely with data, platform, and application teams to build secure, scalable AI workloads leveraging Red Hats Kubernetes-based stack, open-source tooling, and modern MLOps/GenAI Ops practices.

This role defines the technical vision, learning roadmap, reference architectures, and reusable solution patterns required to help internal teams and customers adopt, operationalize, and scale AI and generative AI platforms across hybrid cloud environments.

Key responsibilities :

- Design and implement AI/ML and GenAI architectures on Red Hat OpenShift AI and Red Hat AI Enterprise, including training, tuning, inference, and agentic workflows.

- Build and maintain AI/ML pipelines (data prep, training, evaluation, serving, monitoring) using containerized workloads and Kubernetes-native tooling.

- Integrate models and applications with the broader Red Hat stack (RHEL, OpenShift, automation, observability) and hardware accelerators such as GPUs.

- Implement high-performance inference solutions, including LLM/RAG, model compression/quantization, and scaling patterns across clusters and hybrid clouds.

- Collaborate with data engineers, data scientists, and application teams to productionize AI use cases and ensure reliability, security, and compliance.

- Contribute to MLOps/GenAI Ops practices: model lifecycle management, CI/CD for ML, monitoring, drift detection, and retraining strategies.

- Evaluate and integrate open-source AI frameworks and partner technologies into the Red Hat AI ecosystem, creating reference architectures, demos, and proofs of concept.

- Provide technical guidance, standards, and best practices for building AI-native and agentic applications using Red Hat AI capabilities and unified APIs.

Required skills and experience :

- Deep expertise in Kubernetes and Red Hat OpenShift, including platform architecture, security, scalability, and hybrid cloud deployment patterns.

- Strong understanding of AI/ML and generative AI workflows, including data preparation, model training, fine-tuning, inference, evaluation, and monitoring.

- Experience developing reference architectures and reusable design assets for enterprise technology adoption.

- Experience designing and operating data and ML pipelines, including training, batch/online inference, and monitoring in production environments.

- Hands-on development experience in Python or Go, with practical application of machine learning or deep learning frameworks (e.g., PyTorch, TensorFlow).

- Practical familiarity modern AI/ML ecosystems, including LLM-related patterns such as retrieval-augmented generation, agent-based workflows, and inference optimization.

- Hands-on experience with Red Hat OpenShift AI, including model serving, data science pipelines, model monitoring, and distributed workloads.

- Familiarity with Red Hat AI Enterprise capabilities such as integrated lifecycle management, high-performance inference, agentic AI workflow management, and hybrid deployment consistency.

- Exposure to Red Hat Enterprise Linux AI and Granite-related learning or deployment patterns.

- Familiarity with cloud platforms (AWS, Azure, GCP) and hybrid cloud patterns, especially in combination with OpenShift-based platforms.

- Establish best practices for MLOps and GenAIOps, including data science pipelines, observability, performance tuning, distributed.

- Excellent communication skills and ability to work with cross-functional stakeholders (leadership, data science, engineering, operations).

- Experience with Red Hat OpenShift AI (RHOAI) specifically, or similar enterprise AI platforms.

- Familiarity with Red Hat AI Enterprise concepts such as unified AI/agentic workflow management and Llama-based stack integrations.

- Experience tuning AI performance and scale, including GPU optimization, distributed training, and large-scale inference.

- Designing or advising on vLLM-based inference architectures for low-latency and high-throughput LLM serving.

- Understanding how llm-d extends vLLM through inference-aware scheduling, disaggregated prefill and decode serving, and multi-tier KV-cache strategies for distributed inference.

- Evaluating when standalone vLLM is sufficient versus when enterprise-scale deployments require llm-d for elasticity, cache-aware routing, and Kubernetes-native orchestration.

- Applying these patterns within OpenShift or Kubernetes environments to support scalable generative AI and agentic AI workloads.

Success measures :

- Increased internal and customer readiness to design and deploy Red Hat AI solutions through structured enablement programs and reusable artifacts.

- Faster time to architecture adoption by standardizing validated patterns for AI platform design, inference, governance, and operations.

- Higher quality and consistency in customer implementations through reference architectures, workshops, and field-tested best practices.

- Stronger strategic positioning of the organization as a trusted advisor for Red Hat AI platform adoption and expertise development.