Ditch Legacy Cloud: Orchestrate Edge AI Instead

Why Edge AI Can't Wait: The Race to Deploy Intelligence Outside the Cloud

What happens when the cloud can't reach you? In Episode 13 of the Edge Tech Podcast, we sit down with Said Ouissal, CEO and Founder of ZEDEDA, to explore why edge AI computing is no longer a future concept. It's already running in car washes, on container ships, at airport security checkpoints, and inside oil fields. Said shares how ZEDEDA identified the convergence of cloud-native development and the hard limits of cloud connectivity nearly a decade ago, and built an orchestration platform that brings cloud simplicity to edge deployments at scale, without compromising on security or performance. We go deep on: * Why AI at the edge is outpacing data center AI for real-world industrial and enterprise use cases * The security risks nobody talks about including physical hardware theft, AI model theft, and prompt injection at the edge * ZEDEDA's new Edge Intelligence Platform (EIP) and how it simplifies AI inference deployment across up to 1,000 edge nodes with zero manual optimization * Real deployments across retail, industrial, and critical infrastructure, including Sonny's Car Washes, Maersk, and Volkswagen * Said's bold prediction: within 12 to 24 months, most new applications won't be traditionally written software. They'll be AI agents Whether you're managing edge infrastructure across a manufacturing floor, a federal network, or hundreds of QSR locations, this episode is essential listening.

Ditch Legacy Cloud: Orchestrate Edge AI Instead

Legacy cloud wasn't built for the edge and AI is exposing that gap fast. In this episode, David Kosloski sits down with Padraig Stapleton from ZEDEDA to make the case for ditching legacy cloud architecture and replacing it with purpose-built edge AI orchestration because when decisions need to happen in milliseconds, sending data back to a data center is no longer an option. From manufacturing floors to energy sites, enterprises are moving beyond experimentation. But most edge computing projects fail before they ever reach production. Padraig breaks down the real reasons why security gaps, operational complexity, and the painful truth about building in-house vs. deploying enterprise-grade platforms. In this episode, we cover: Why AI infrastructure at the edge requires cloud-like orchestration and why most teams underestimate that The rise of agentic AI and small language models powering real-time autonomous decision-making How edge AI is showing up right now in quality inspection, safety monitoring, and autonomous systems The security non-negotiables that every edge deployment must clear What it actually takes to scale from a 10-device pilot to thousands of distributed endpoints Why AI will eventually become invisible and what that means for the businesses deploying it today Whether you're an engineer evaluating edge AI platforms, an enterprise architect navigating AI infrastructure decisions, or just tracking where the industry is heading — this conversation is one you don't want to miss. 🔗 Learn more about Edge AI: SNUC.com [http://snuc.com] ⏱ Chapters: 0:00 AI shift from cloud to edge 1:05 Intro + guest background 3:10 Why edge orchestration matters 4:20 Edge vs data center challenges 5:40 Why edge projects fail at scale 7:00 Security and enterprise barriers 9:00 Build vs buy at the edge 12:15 Agentic AI and future architectures 14:00 Real-world AI use cases 15:30 AI in the physical world (robots, drones) 16:30 The future of invisible AI 18:00 Innovation and closing thoughts

The Future of Edge Infrastructure: Fast Kubernetes Deployments Explained

In Episode 11 of the Edge Tech Podcast, host David Kosloski sits down with Brian Durden, VP of Solutions Engineering & Strategic Initiatives at Rancher Government Solutions, to discuss why speed and sustainability are becoming critical for Kubernetes deployments at the edge. As hardware becomes more powerful and software more efficient, organizations can now deploy complex mission applications directly at the edge without relying on constant cloud connectivity. But speed isn't just about performance — it's about how quickly platforms can be built, deployed, and maintained in real-world environments. In this conversation, we explore: * Why Kubernetes has become so dominant in edge infrastructure * The three types of speed that matter for edge deployments: build, deploy, and runtime performance * Why sustainable edge platforms are critical for government and mission-critical environments * How automation and standardization reduce operational complexity * The role of compact edge hardware in enabling distributed deployments Brian also shares insights from years of experience building platforms across VMware, Pivotal, and Rancher Government, including how organizations are rethinking infrastructure to support distributed, disconnected, and low-bandwidth environments. If you're building edge platforms, distributed infrastructure, or Kubernetes environments, this episode dives deep into the real-world challenges and opportunities shaping the future of edge computing.

Addressing Compliance Bottlenecks in Defense AI: The ATO-in-a-Box Approach

Mission-ready edge computing isn't the same as "traditional" edge. In this episode of the Edge Tech Podcast, host David Kosloski sits down with Tommy Scherer (Principal Architect) and Mark Perry (Director of Public Sector Growth) at Spectro Cloud to break down what makes tactical edge different—running mission apps in contested, disconnected environments with security and compliance built in from day zero. We cover: * What "tactical edge" really means (hint: secure by default) * Zero Trust, encryption, and operating as if the network is already compromised * Why mission workloads (ISR, secure messaging, computer vision) need edge platforms that can deploy fast and evolve continuously * How ATO and continuous ATO help accelerate fielding new capability * Why the future is about portable compute + hardened stacks that look and operate like the data center If you're building for defense, federal, or regulated environments, this one's for you.

Streaming "Impossible" Video: Solving Cellular Limits at the Edge

How do you stream reliable, real-time video when the network is congested, unstable, or barely there? In this episode of the Edge Tech Podcast, David Kosloski sits down with Clive Sawkins (COO) and Dale Hall (Head of Product Engineering) from Digital Barriers to answer that exact question. We cover the mechanics of adaptive video codecs and explore how moving away from dedicated hardware is redefining video surveillance and situational awareness. From congested festivals like Glastonbury to remote border security, we discuss how to maintain mission-critical connectivity anywhere. 🚀 Key Topics Covered: The Bandwidth Bottleneck: Why traditional cellular video fails when you need it most. Intelligent Compression: How Digital Barriers' codec mimics human perception to save data. The Invisible Edge: Why the best edge ai technology works without you noticing it. Beyond Security: Using video as a data sensor for agriculture, telemedicine, and autonomous systems. ⏱️ Timestamps: 00:00 - Start 01:02 - Podcast Intro: Welcome & Guest Introductions 01:21 - Video Over Cellular: The struggle with LTE and 5G 01:34 - Bandwidth Challenges: Uplink limits & constraints 03:03 - Digital Barriers Tech: Adaptive video compression 04:25 - AI vs Machine Learning: Intelligent video delivery 05:24 - What Is the Edge? Defining edge computing today 06:28 - Border Security Use Case: Surveillance as infrastructure 07:10 - Advanced Camera Support: Thermal, Night Vision, PTZ 07:46 - Rapid Deployments: Mobile & Emergency setups 08:12 - Festival Case Study: Reliable video in crowds 08:53 - Why Phones Fail: Consumer vs. Mission Critical 09:46 - Network Redundancy: Cellular + Satellite 11:07 - Video as a Sensor: AI Analytics 12:12 - Software-Only Model: Deployment flexibility 13:39 - Hardware Independence: Freedom at the edge 14:02 - Video as Data: Driving AI insights 15:08 - Event-Based Streaming: Triggered video 15:36 - Sensors & Power Efficiency: Low-power design 17:03 - Industry Applications: Security & Utilities 18:23 - Unexpected Use Cases: Innovation 19:02 - Extreme Environments: Harsh conditions 20:08 - Invisible Edge Tech: When it just works 21:27 - Closing Thoughts