WWT Sets a New Standard for Enterprise-Ready AI with Cisco Secure AI Factory with NVIDIA
In this blog
Enterprises are under pressure to move AI out of the lab and into daily operations. But turning early pilots into production systems that can perform reliably, securely and at scale remains one of the hardest challenges in modern IT. Many organizations can assemble the pieces. Far fewer can make them work together once real workloads and real users hit the system.
WWT, Cisco and NVIDIA are addressing that gap with the Cisco Secure AI Factory with NVIDIA — an enterprise architecture designed by Cisco and NVIDIA and operationalized by WWT for real-world environments. The result is not just a reference design, but a practical foundation for running AI in production.
Neil Anderson, WWT Vice President of Cloud, Infrastructure and AI Solutions, and Jeff Fonke, WWT Practice Director of High-Performance Architecture, share how this collaboration gives customers a clear path from AI ambition to AI systems that can stand up to enterprise demands.
Solving what's hard about enterprise AI, together
The hardest part of enterprise AI is not choosing models or hardware — it's making everything work together once the system is live. AI development and delivery software, production AI-powered applications, data pipelines, accelerated compute, high-performance networking, and security controls all have to operate as a single, reliable platform under real production pressure.
That is the problem WWT, Cisco and NVIDIA set out to solve with the Cisco Secure AI Factory with NVIDIA.
"Our job is to take the complexity out of getting AI to production," said Anderson. "Customers don't just want AI. They want AI that works."
That's where the combined strengths of all three organizations become essential.
Cisco's leadership in AI-ready networking and embedded security gives organizations a foundation they can trust as they scale.
NVIDIA continues to push the boundaries of accelerated computing with the platforms and inferencing microservices that power today's most demanding AI workloads.
WWT brings practical integration know-how and global delivery capabilities, helping organizations cut through complexity and adopt these technologies in ways that genuinely move their business forward.
According to Fonke, "Our depth in NVIDIA AI Enterprise software and inferencing is based on early adoption of NVIDIA Blueprints and NVIDIA NIM, continued collaboration with NVIDIA's software experts, and real hands-on experience. We're helping architect the network backbone modern AI systems depend on, working side by side with Cisco to refine designs that operate at a true enterprise scale. And as the first partner to deploy Cisco AI Defense on Cisco Secure AI Factory with NVIDIA, we're proving how advanced security, networking and AI operations come together in real environments, and not just on paper."
"When our teams come together, we can offer something customers can trust: a secure, scalable AI foundation built for real production demands. Working together has allowed us to build an approach that's far more impactful than any one of us could have created alone, and we're proud of that," Anderson added.
The Cisco Secure AI Factory with NVIDIA delivered by WWT
The Cisco Secure AI Factory with NVIDIA is a modular reference design that accelerates the delivery of trusted and transformative AI applications. This integrated offering combines Cisco's market-leading networking, security, observability and compute capabilities with NVIDIA's cutting-edge AI software, accelerated compute and networking, along with third-party storage certified by both NVIDIA and Cisco.
WWT operationalizes this design by turning the modular reference design into a production-ready platform that customers can run at enterprise scale.
WWT's more than 35 years of experience in deploying complex, integrated solutions globally — combined with its AI Proving Ground; robust AI Studio, AI Foundry and AI Factory services; and its AI Readiness Model for Operational Resilience (ARMOR) framework with NVIDIA AI — help customers derisk adoption, align architecture to business needs and deploy AI at scale with confidence.
"Our Proving Ground, ARMOR framework and scale create a clear advantage," Fonke said. "Together with Cisco and NVIDIA, we're helping customers build AI systems that perform reliably in production."
What sets WWT apart in operationalizing the Cisco Secure AI Factory with NVIDIA
This collaboration gives customers a clear, confident path to operational AI, built on three realities:
1. Enterprises need an AI factory that suits their needs and aligns with their business goals to leverage AI at scale.
Many organizations have been pitched AI "factories." Few have seen one function end-to-end in their own environments. This solution is backed by real deployment experience, real architectural refinement in WWT's AI Proving Ground and real customer outcomes.
2. Scale is where most AI programs break down, and where WWT excels.
From hyperscale deployments to heavily regulated industries, WWT teams have identified architectural gaps early, prevented costly redesigns and ensured systems hold up under real-world load.
"Real deployment work reveals gaps quickly," Fonke noted. "We've helped customers avoid issues that could have slowed them down significantly."
3. The AI Proving Ground removes the biggest uncertainty: "Will this work for us?"
Inside WWT's AI Proving Ground, customers can test the architecture, trial secure pipelines, and understand what model refinement, agentic workflows and scaled inference look like before committing resources.
WWT's AI Studio and AI Foundry extend this ecosystem's capabilities by helping organizations assess use cases and align technical decisions with business outcomes.
The AI Proving Ground packages what enterprises need to adopt AI so teams can try now and keep building as resources grow: hands-on access, curated learning and working blueprints.
Hands-on labs
These labs create space for teams to experiment with the Cisco Secure AI Factory with NVIDIA's architecture, including NVIDIA AI Enterprise software, Cisco's high-performance compute and networking stack, and the surrounding operational components, including cluster management and security integrations.
Several labs are already available, with additional labs actively in development.
The following capabilities will be incorporated into the upcoming lab experiences:
- NVIDIA AI Enterprise software, NVIDIA NIM, NVIDIA NeMo and NVIDIA Run:ai integration
- Security capabilities around Cisco's AI Defense and WWT's ARMOR framework
- Cisco Splunk Observability features
- Cisco UCS C and X-Series servers based on NVIDIA HGX™ and NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPUs, including foundational factory operations including foundational factory operations including foundational factory operations
- Enterprise high-performance storage integrations and interoperability
Targeted learning paths
Learning paths are guided education that helps infrastructure, security and data science teams work together to adopt and scale AI using a unified platform rather than siloed tools, with additional Cisco Secure AI Factory with NVIDIA learning paths are actively in development.
A working use case: NVIDIA AI Blueprint for Building Data Flywheels on the Cisco Secure AI Factory with NVIDIA
A practical example of this architecture in action is WWT's implementation of the NVIDIA AI Blueprint for building data flywheels within the Cisco Secure AI Factory with the NVIDIA environment in the AI Proving Ground.
The NVIDIA AI Blueprint for building data flywheels is designed to continuously improve AI models using real-world data. It starts with a large, high-accuracy model running in a central environment, captures usage data and difficult examples, retrains and fine-tunes the model, and then produces smaller, optimized models that can be deployed closer to where work happens.
WWT runs this core process on Cisco infrastructure and then generates right-sized models for edge environments. This reduces latency and hardware requirements while maintaining model accuracy.
To demonstrate this in practice, WWT is building a prototype medical coding suggestion application that depends on a large base model for accurate recommendations. Using the Data Flywheel Blueprint, WWT can show how the system identifies a smaller, tuned model that delivers nearly the same inference results at a lower cost and with faster response times.
This approach delivers two tangible benefits: lower inference costs through reduced GPU requirements and the ability to run models closer to users and data, improving performance and responsiveness.
"The NVIDIA AI Blueprint for building data flywheels featured on the Cisco Secure AI Factory with NVIDIA architecture will serve as a practical guide for customers who want to see how continuous model tuning and data refinement can work in an enterprise setting," said Fonke. "We're excited to demo these capabilities at NVIDIA GTC as they mature."
Anderson concluded, "For us, the most rewarding part of this work is seeing customers move from AI ambition to AI outcomes and knowing the architecture underneath them is strong enough to support what comes next."
See it firsthand at NVIDIA GTC in San Jose
We're excited to bring these advancements to NVIDIA GTC, March 16–19, 2026, in San Jose, where we'll be showcasing:
- Live demonstrations of the Cisco Secure AI Factory with NVIDIA
- Working AI Proving Ground demos, including the NVIDIA AI Blueprint for building data flywheels use case
- Opportunities to sit down with experts and dive deeper into design strategy, security and operational readiness
If you're planning to attend, stop by booth #1821! This is a great opportunity to explore what WWT, Cisco and NVIDIA have been building together and to get a firsthand look at where we're headed next.
Register and learn more about WWT's presence at NVIDIA GTC.
Also joining Cisco Live EMEA in Amsterdam
We'll also be on site at Cisco Live EMEA, February 9–13, 2026, in Amsterdam.
Visit us at booth #E03 to connect with our experts and explore how we're helping organizations accelerate secure, enterprise-ready AI initiatives alongside Cisco.