Unlocking agent-native transformation with CrewAI Factory and NVIDIA

Day 1 of CrewAI Launch Week 01 (May 2025)

CrewAI Enterprise makes it easy for organizations to build, run and monitor Agentic AI workflows in the cloud – and to automate business processes on day one. However, in practice, large enterprises often need to run AI applications on their own infrastructure due to security and data privacy requirements.

That’s why we’re excited to announce CrewAI Enterprise is now available for deployment on customer infrastructure via CrewAI Factory – whether in a data center or a virtual private cloud. CrewAI Factory provides the same benefits and capabilities of CrewAI Enterprise, but in a customer-managed form factor that provides:

• Management and monitoring
• Real-time testing and tracing
• No-code development
• Enterprise tools and connectors
• Encrypted communication and storage
• Secure operations (e.g., SOC2/HIPPA/FedRAMP High)

Deploy anywhere

AWS/Azure VPCs – AWS Fargate and Azure Container Apps support enables enterprises to run on their own VPCs and with their own cloud databases (e.g., AWS RDS/Azure Database).

Kubernetes clusters – Kubernetes deployment via Helm enables enterprises to run on their own AWS Elastic Kubernetes Service, Azure Kubernetes Service and Google Kubernetes Service clusters as well as Red Hat OpenShift (on premises or in the cloud).

Serverless scaling, high availability – A serverless, container-based architecture not only simplifies deployment, but provides automatic scaling via resource allocation and high availability via persistent volumes and database replicas.

Secure authentication – Native integration with Auth0 and Microsoft Entra ID for customer-managed, enterprise-wide authentication.

Become agent-native

CrewAI Factory not only helps enterprises meet security and data privacy requirements for AI applications, it bridges the gap between CrewAI’s agent-powered and agent-native adoption levels. It’s one thing to deploy a handful of agentic workflows and see the potential for massive impact on the business. It’s another to embrace Agentic AI as means to fundamentally transform business processes and accelerate growth.

With CrewAI, enterprises can scale several orders of magnitude – a dozen agentic workflows vs. hundreds if not thousands, a handful of simple use cases vs. hundreds of critical use cases spanning multiple teams, departments or business units.

The path to becoming agent-native is the real promise of CrewAI Factory. Beyond a framework for orchestrating agents, it’s a complete platform designed to infuse Agentic AI into the DNA of innovative enterprises.

Build an AI factory

CrewAI worked with NVIDIA to help enterprises with this transformation, with CrewAI Factory providing the Agentic AI platform and NVIDIA AI Enterprise providing specialized models and tools for agents via NVIDIA NeMo and NVIDIA NIM microservices.

The integration of NIM microservices enables CrewAI customers to build and deploy agents built on NVIDIA NeMo Retriever, NVIDIA Llama Nemotron and NVIDIA NeMO Guardrails models. With access to these models and secure, high-performance inference, agentic workflows on CrewAI Factory benefit from advanced reasoning and tool use as well as higher accuracy when retrieving knowledge from unstructured data. NVIDIA hosts a number of reference AI blueprints, with one for code documentation submitted by CrewAI here.

The deployment of CrewAI Factory and NVIDIA AI Enterprise provides customers with their own AI factory, running on their own infrastructure and with no limits on scale and specialization.

Agentic AI is empowering organizations to innovate faster in virtually every industry, CrewAI and NVIDIA are working together to empower enterprises to orchestrate intelligent, multi-agent systems that reason, plan, and act, scaling productivity through intelligent AI teammates.”

John Fanelli, VP of Enterprise Software at NVIDIA.

CrewAI's first launch week is just getting started. Stay tuned as we have a lot more to share, and sign up for our webinar this Thursday to discover how we're taking Agentic AI to the next level!