Surviving Scale: Building AI Workloads That Hold Up Under Demand

The Power of Broke: Innovation Lessons from Daymond John

Innovation does not have to start with unlimited resources. For Daymond John of ABC’s Shark Tank, it starts with the power of broke. In this special episode of Edge of IT, recorded live from Pellera’s West Tech Summit, Jonathan sits down with entrepreneur, investor, and bestselling author Daymond John to explore risk, trust, and innovation under pressure. From building FUBU in his mother’s kitchen to evaluating hundreds of founders on ABC’s Shark Tank, Daymond shares why success is rarely about having the perfect idea. It’s about understanding the problem, surrounding yourself with the right people, managing your downside, and doing the work before asking others to believe. Together, Jonathan and Daymond connect those lessons to enterprise technology today, where AI is moving fast, pressure is high, and leaders are being asked to make smart bets without perfect certainty.

AI Agents in Practice: Building Systems People Trust

A chatbot is easy to launch. A chatbot that earns trust, improves experience, and proves ROI? That is where things get interesting. In this episode, host Jonathan Gough explores what it really takes to build AI agents that people actually want to use. Joining him are Matt Grayson, CIO and IT Division Manager at APWU Health Plan, and Theo Pajaczkowska, Advisory & Design Managing Principal for Data & AI at Pellera. Together, they unpack how Matt’s team, in collaboration with Pellera, built a successful virtual agent on IBM WatsonX Assistant and watsonx AI to improve the self-service experience for members navigating health insurance questions during open enrollment. Operating in the highly regulated healthcare space, Matt shares how his team balanced user experience with governance, while Theo adds her valuable perspective on Pellera's outcome-driven approach for designing vendor-agnostic AI solutions. So if you want to build AI that people use, trust, and come back to, start here.

Adaptive AI Strategy: Balancing Innovation and Experimentation

As AI accelerates, many organizations are realizing that the old rules of strategy no longer hold any weight. In this episode of Edge of IT, Jonathan Gough is joined by Kree Govender, SMB Canada Lead at Microsoft, and Jonathan Real, Solution Architect at Pellera Technologies, to explore what AI readiness looks like when change is constant. Together, they unpack why long-range roadmaps are breaking down, and why the organizations pulling ahead are the ones creating adaptable strategies, faster feedback loops, and better paths from idea to execution. The conversation also digs into how to encourage AI experimentation without losing control. From shadow AI and psychological safety to governance, guardrails, and data readiness, Kree and Jonathan offer a look at how leaders can move faster with AI while still staying secure.

AI in the Enterprise: Governance, ROI, and the Push for Forward Compatibility

As teams experiment with new tools and leaders push for AI-first thinking, many organizations are discovering that AI for AI's sake is just not cutting it.  In this episode of Edge of IT, Jonathan Gough sits down with Demetrios Brinkmann, founder of MLOps Community, to explore the growing tension between personal AI use and organizational AI adoption. Together, they unpack why AI ROI remains so difficult to measure, where value is actually showing up today, and why simply layering AI onto existing processes can create more friction than progress. Jonathan and Demetrios also dig into the deeper questions emerging beneath the surface: how governance must evolve alongside capability, why coding agents and support workflows are becoming early proving grounds, and what it means to build with “forward compatibility” in mind as models, costs, and infrastructure continue to shift. If you’re trying to understand not just what AI can do, but how to adopt it without losing trust, clarity, or strategic direction, this episode is for you.

Surviving Scale: Building AI Workloads That Hold Up Under Demand

First question: Are you designing an AI workload? Second question: Are you ready to respond when your team asks you to scale it?  As more small AI use cases emerge, organizations are feeling pressure to scale, and big questions are being raised from technical, operational, and cost perspectives. In this episode of Edge of IT, Jonathan Gough speaks with Robbie Jerrom, Principal Technologist for AI at Red Hat, and Josh Minthorne, Director of Cloud, DevSecOps & Automation at Pellera, about what it takes to “raise” AI from an early win into a scalable, governed workload. Robbie, Josh, and Jonathan share personal stories from the trenches, navigating the inevitable challenges that come with scaling AI workloads. They unpack why demand is so hard to predict, why startup times and GPU requirements can surprise you, and how internal adoption can spike fast enough to break systems. So, if you’re wondering how to adapt, this episode outlines the next steps your team needs to take to get back on track.