Host: Cedric Clyburn
Guest: Jehlum Vitasta Pandit and Jenny Yi
Producer: Rohan Venkatram
Follow along: https://www.redhat.com/architect/portfolio/detail/137-whats-new-openshift-ai-interactive-experience?intcmp=RHCTG0250000449876
👉 What you’ll see in this episode:
🔹 What’s New in Red Hat OpenShift AI 3.3 – From Pilot to Production
🔹 AI hub – Centralized assets for enterprise AI
Managing model discovery, versioning, and deployment insights
Performance guidance from Red Hat AI model validation program
Supporting the future of multimodel and multiagent AI systems
🔹 Governance at Scale with Model-as-a-Service (MaaS)
Technical preview of MaaS capabilities in OpenShift AI 3.3
Delivering centralized AI models to developers via API endpoints
Role-based access policies and usage governance
Rate limiting policies and usage quotas in the UI
Optimized inference routing with llm-d for efficient GPU utilization
🔹 Accelerating Developer Velocity with Gen AI Studio
Technical preview of Gen AI Studio developer environment
AI playground for prompt experimentation and model tuning
Importing and managing MCP tools for agentic workflows
“View Code” capability for exporting configurations to local development
Quick access to API endpoints and keys for testing in developer environments
🔹 Closing the Production Gap with Continuous Evaluation
Cost optimization with LLM compression and benchmarking
Guided workbenches for LLM Compressor and GuideLLM
Tracking experiments and model performance with MLflow integration
Visualizing performance improvements and inference latency correlations
🔹 Interactive Q&A with Cedric, Jehlum & Jenny
Interactive Demo / Reference:
https://www.redhat.com/en/blog/whats-new-red-hat-openshift-ai-33-ui-moving-pilot-production
🎯 Episode Overview:
Cedric Clyburn returns with another Demo Deep Dive, this time joined by Jehlum, Product Manager for Red Hat AI, and Jenny, Technical Product Manager for Red Hat AI Enterprise.
Jehlum focuses on building platforms for generative AI applications, with particular interest in data processing, observability, safety, and evaluation—key components for building production-grade generative AI systems at scale.
Jenny brings experience designing end-to-end platform experiences for Red Hat AI Enterprise. Prior to joining Red Hat through the Neural Magic acquisition, she developed interfaces for LLM benchmarking and AI control planes. Her background consulting for healthcare and public health organizations informs her focus on building AI tools for high-stakes, highly specialized domains.
In this episode, we explore the latest innovations in Red Hat OpenShift AI 3.3 and how the platform helps organizations move generative AI projects from experimentation to production. The session will showcase new capabilities including the AI hub for centralized model management, Model-as-a-Service (MaaS) governance, Gen AI Studio for developer experimentation, and new optimization and evaluation tools designed to help teams manage cost, performance, and reliability in enterprise AI deployments.
Whether you’re building internal AI platforms, enabling developer experimentation, or operationalizing generative AI at scale, this episode demonstrates how OpenShift AI 3.3 bridges the gap between pilot projects and production-ready AI systems.
🔗 Learn more and interact:
📌 Explore Red Hat OpenShift AI – https://www.redhat.com/en/technologies/cloud-computing/openshift/openshift-ai
📌 What’s new in OpenShift AI 3.3 – https://www.redhat.com/en/blog/whats-new-red-hat-openshift-ai-33-ui-moving-pilot-production
📌 Try Red Hat OpenShift AI – https://www.redhat.com/en/technologies/cloud-computing/openshift/openshift-ai/trial
📌 Follow Cedric – https://www.cedricclyburn.com/
📅 Livestream: Tuesday, March 10th @ 11am ET
▶️ Watch live or on-demand on @RedHat and @OpenShift
#RedHat #OpenShiftAI #GenerativeAI #EnterpriseAI #LLM #AIGovernance #AIPlatform #HybridCloud #Kubernetes
