Exquisite Alignment

Hewlett Packard Enterprise gathers the industry’s best to make GenAI applications safer, faster, and easier to deploy

Call it kismet. One of the most important research papers in deep learning was published on June 12, 2017. Attention is All You Need introduced a blueprint for building AI models that worked with sequences of data to power large language models (LLMs) such as ChatGPT and other chatbots. That transformer reshaped the AI market – and the world.

Chad Smykay

“It’s funny that we’re talking about this today on the eighth anniversary of the Transformers Archive paper.” The “this” Chad Smykay is referring to is an extraordinary partnership between Hewlett Packard Enterprise (HPE), Mark III Systems, and GambitCo to develop landmark solutions for securing intellectual property and data used in AI for the enterprise, streamlining time to market, and keeping costs controlled. Their combined strength complements a longstanding collaboration between HPE and NVIDIA to deliver co-developed enterprise solutions and joint go-to-market integrations to help businesses succeed with AI.

As the Chief Technology Officer for AI and Distinguished Technologist for HPE, Smykay supports all of their verticals, including manufacturing, energy, retail, healthcare, life sciences, and several others.

For years prior to joining HPE, Smykay worked on early machine learning models and distributed frameworks like AlexNet, a neural network architecture developed for image classification tasks, and the Apache MXNet deep learning framework for developing and deploying deep neural networks across various platforms.

Bringing such real world, hands-on experience to the tech table gives HPE a core advantage in a maturing market.

“How you think, how you run your business, all of that’s getting leaked out to AI right now,” he warned. When information is put into a generative AI chatbot such as ChatGPT, it stays there and gets reused to retrain their models. “As we see these technologies being adopted by customers across all industry verticals, security around the data and intellectual property, and what’s out there, is becoming more and more important.”

AI is a powerful accelerant for most businesses, but development costs can quickly throttle such plans.

Today’s chat models aren’t just spitting out responses to queries, they’re “thinking” through problems, which increases the cost of the technology. “You need a fixed cost model, and that’s another thing that HPE brings to the table,” he noted. Fixed costs mean speedier development, tighter data privacy, and a way to control spending.

Partnership Power

To bring their ideas to fruition, HPE and Smykay are journeying with partners to create AI solutions that deliver unmatched speed, precision, and scale.

Digital and IT transformation solutions provider Mark III Systems takes a unique “full stack” approach to modernizing and optimizing core tech stacks in the data center, but also spin up and scale data-driven, digital transformation stacks in the data center, in the cloud, and on the edge.

GambitCo brings lightning speed to the project; the blending of their cutting-edge AI solutions with HPE’s Private Cloud AI infrastructure and Mark III Systems’ integration and deployment expertise, they are dramatically reducing the time it takes for organizations to realize value from AI.

Pre-integrated solutions on HPE Private Cloud AI mean faster onboarding, deployment, and scalability. Mark III’s proven integration playbooks shorten deployment timelines from months to weeks or even days. Customers get immediate access to turnkey AI capabilities built for real-world workflows across industries — no heavy lifting required — allowing AI to be a force multiplier for human action across verticals.

“GambitCo, Mark III Systems, and HPE are redefining how fast AI can deliver real-world value,” said Chris Bogan, Mark III VP of Sales. “With our unified approach, the sky isn’t the limit. GambitCo, Mark III, and HPE Private Cloud AI Solutions are the launchpad.”

The AI Buzz

If you visit HPE’s Customer Innovation Center in Spring, Texas, get ready for the bees. The with military veterans to care for the bees. Not only does the tending help vets with PTSD, it demonstrates the intriguing impact of AI in real life through nature.

“We’ve discovered that artificial intelligence can be used across every industry for just about every kind of use case … it can be applied anywhere, and it just helps people visually see it.”

“The bees are equipped with cameras and acoustic microphone sensors,” Smykay said, noting that DARPA is doing amazing research on bees as well. “There’s research that suggests that we can find out what kind of diseases bees have based on the sound. So we’re like, ‘Cool. Let’s see if we can detect that.’” Work has only just begun on that project. “We’ve been monitoring with visuals in temperature sensors for the last three years.”

One study examines how bees protect themselves in winter. “We find out how they swarm into certain hives that are established with a queen, how many are already in there,” he said. “We’re using visual language model tools from partners like Iron Young, a great partner of ours called Vadio now. We’re literally counting the bees just through visuals. It works well, so we’re using this program on many other things.”

In the future, Smykay envisions selling their honey and giving the proceeds back to nonprofits around the Houston area. “Of course, we’re enabling technology and sharing that data with researchers out in the greater bee community. It’s been a lot of fun to work on.” HPE is continuing to grow and scale the effort with technology and additional partnerships.  The program shows the utility and versatility of AI. If you can use it here, you can use it anywhere.

Sharing the knowledge wealth

Taking the guesswork out of deploying AI solutions is a key aspect of their collaboration. “There are great open-source software alternatives. There’s the best of breed from Nvidia’s AI Enterprise software with our co-partnership and engineering efforts with NVIDIA and HPE. The ease of deploying GambitCo is cookie-cutter. … We take out all the guesswork of setting these things up.”

Throughout his career Smykay has collected plentiful advice for those considering their own AI adventure. “Make sure you have an open technology stack so you can integrate with multiple models, AI agents, and data sources. Don’t hoard that data just for your use case, keep it open for future growth.” he instructed.

“You also need to make everyone aware of what you have so you can reuse it for other types of specific use cases in your organization. If you’re starting your AI journey, have a buying committee defined. Don’t have one person doing it, not just somebody in technology, not just somebody just in business, not just somebody that’s just a data scientist doing some ‘cool’ thing without relating it back to the business.

“To relate it back to business goals, you’ve got to put a committee together so everyone knows what the impact on your business could be, as well as how you plan to support and implement it. If you do not know where to start in your organization, that’s where you bring in folks like HPE and Mark III to help fill in those gaps.” It’s guaranteed to be a match made by the brightest stars.