The Edge Computing Podcast with Fexingo: Local Compute, CDNs, and Distributed Infrastructure

Lucas and Luna explore how edge computing is solving the latency problem that has always plagued cloud gaming. They focus on NVIDIA's GeForce NOW and its deployment of edge nodes in partnership with telcos like SoftBank and Deutsche Telekom, reducing input lag from 150ms to under 20ms. The hosts discuss why traditional CDNs weren't enough, the role of GPU virtualization, and how edge nodes handle real-time frame rendering and streaming. They also touch on the business model — subscription versus per-session pricing — and what this means for the future of AAA gaming without expensive hardware. A concrete look at how distributed infrastructure is turning game streaming from a nice idea into a real product.

#EdgeComputing #CloudGaming #NVIDIA #GeForceNOW #Latency #Technology #GPU #Virtualization #SoftBank #DeutscheTelekom #CDN #GameStreaming #FexingoBusiness #BusinessPodcast #DistributedInfrastructure #RealTimeRendering #InputLag #AAA

Keep every episode free: buymeacoffee.com/fexingo

Episode 97 of The Edge Computing Podcast explores how major insurance carriers are deploying edge servers at broker offices and auto shops to run real-time risk models. Lucas and Luna walk through a specific use case: a regional carrier that cut quote-to-bind time from eight minutes to under 30 seconds by moving AI inference from a central cloud to a local edge node. They discuss the hardware choices — NVIDIA Jetson and Intel Xeon D — the data-privacy advantages of keeping customer information local, and the resulting loss-ratio improvement. The episode also touches on how this model is starting to spread to workers' comp and property inspection. A concrete look at a quiet but massive edge deployment that most consumers never see.

#EdgeComputing #InsuranceTech #RealTimeUnderwriting #AI #NVIDIAJetson #IntelXeonD #RiskModeling #DataPrivacy #BrokerOffice #AutoInsurance #WorkersComp #PropertyInspection #LossRatio #DistributedInfrastructure #FexingoBusiness #BusinessPodcast #Technology #CDNs

Keep every episode free: buymeacoffee.com/fexingo

Episode 96 of The Edge Computing Podcast dives into the real-world deployment of edge-based adaptive traffic signal control. Lucas and Luna examine a pilot project in Phoenix, Arizona, where a 50-intersection corridor reduced average commute times by 27% using on-premise computer vision and low-latency decision-making. They discuss why centralized cloud approaches fail for traffic — latency, bandwidth, and safety — and how edge nodes running lightweight reinforcement learning models can adjust signal timing in milliseconds. The hosts break down the hardware stack: a Jetson-class edge device per intersection, a mesh network between nodes, and a central dashboard for city engineers. They also touch on the regulatory hurdles, the cost per intersection, and how this technology could scale to gridlock cities. No prior edge computing knowledge needed — just an interest in how infrastructure is getting smarter. Tune in for a focused, numbers-driven conversation about one of the most tangible applications of edge computing in civic life.

#EdgeComputing #TrafficSignalControl #SmartCities #Phoenix #AdaptiveTrafficControl #ComputerVision #ReinforcementLearning #Jetson #Latency #MeshNetwork #CivicInfrastructure #Technology #FexingoBusiness #BusinessPodcast #EdgeAI #RealTimeDecisionMaking #UrbanMobility #TransportationTech

Keep every episode free: buymeacoffee.com/fexingo

In Episode 95 of The Edge Computing Podcast, Lucas and Luna explore how edge computing is turning smartphones into pocket-sized supercomputers. They focus on Apple's A-series chips, which now rival laptop CPUs, and the shift from cloud-reliant AI to on-device inference. The episode anchors on the latest A18 chip and its neural engine, capable of running large language models locally. They discuss how this changes app development, privacy, and network demand, with references to Qualcomm's Snapdragon and Google's Tensor chips. A concrete example: real-time language translation that never sends audio to a server. The conversation also touches on the death of the mobile-first paradigm and what edge AI means for the next billion users.

#EdgeComputing #SupercomputerInYourPocket #AppleSilicon #A18Chip #NeuralEngine #OnDeviceAI #QualcommSnapdragon #GoogleTensor #RealTimeTranslation #Privacy #Latency #Tech #Technology #FexingoBusiness #BusinessPodcast #MobileComputing #AIPhone #EdgeAI

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna explore how edge computing is enabling real-time decision-making for drone package delivery. They focus on the specific case of Zipline, the logistics company that uses edge nodes to coordinate autonomous drone flights for medical supply deliveries in Rwanda and the US. Lucas explains how edge computing reduces latency from cloud round-trips to under 10 milliseconds, allowing drones to avoid obstacles and adjust routes dynamically. Luna discusses the computational demands of onboard AI models for object detection and path planning. The hosts also touch on the broader implications for last-mile delivery, including regulatory challenges and infrastructure costs. A concrete example: Zipline's edge nodes process video feeds from drone cameras in real time, enabling safe landing zone identification within seconds. The episode draws on data from the company's 2025 operations, which saw over 1 million autonomous deliveries.

#EdgeComputing #DroneDelivery #Zipline #LastMileLogistics #RealTimeAI #AutonomousDrone #Latency #Technology #TechPodcast #FexingoBusiness #BusinessPodcast #EdgeAI #DroneNavigation #MedicalSupplies #Rwanda #Infrastructure #OnboardCompute #ObstacleAvoidance

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore how edge computing is revolutionizing disaster response by enabling real-time data processing at the network's edge. They dive into the case of Project Owl, a mesh-network system that deployed edge nodes after Hurricane Maria in Puerto Rico, restoring communications within hours instead of weeks. The episode unpacks why centralized cloud fails during infrastructure collapse, how local compute and cognitive radio mesh networks keep first responders connected, and what this means for future emergency preparedness. Specific numbers covered: the 8-hour deployment window versus the 11-month average recovery time for traditional cell towers. A concrete look at how distributed infrastructure is saving lives when milliseconds and local resilience matter most.

#EdgeComputing #DisasterResponse #ProjectOwl #MeshNetworks #FirstResponderTech #PuertoRico #HurricaneMaria #TechForGood #RealTimeData #DistributedInfrastructure #CognitiveRadio #PublicSafety #Resilience #Technology #FexingoBusiness #BusinessPodcast #EdgeTech #NetworkRecovery

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore how edge computing is bringing supercomputer-level performance to mobile devices, using the example of Apple's Neural Engine and on-device AI models. They discuss how distributed inference is reshaping apps from real-time translation to augmented reality, the tension between cloud and edge, and why privacy is a key driver. Specific focus: the A18 Bionic chip's 38 trillion operations per second, and how developers are splitting workloads between device and cloud. A concrete look at the quiet revolution happening in your pocket right now.

#EdgeComputing #OnDeviceAI #AppleNeuralEngine #A18Bionic #DistributedComputing #MobileAI #RealTimeAI #Privacy #AugmentedReality #MachineLearning #InferenceAtEdge #ChipDesign #TechPodcast #FexingoBusiness #BusinessPodcast #Technology #EdgeInference #SmartphoneAI

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore a nascent but critical edge computing use case: real-time spectrum sharing for future 6G networks. They break down the technical challenge of allocating radio frequencies dynamically across thousands of devices without interference, and how distributed edge nodes running lightweight AI models can negotiate spectrum access in microseconds. The episode anchors on a 2025 trial by the University of Bristol and a major European telecom where off-the-shelf edge servers achieved sub-10-millisecond coordination latencies. They also discuss the policy implications, the role of software-defined radios, and why this matters for industrial IoT and autonomous systems. No hype, just the architecture and the numbers.

#EdgeComputing #6G #SpectrumSharing #RealTimeAI #DistributedInfrastructure #Telecom #SoftwareDefinedRadio #UniversityOfBristol #IoT #AutonomousSystems #Latency #Wireless #Technology #FexingoBusiness #BusinessPodcast #EdgePodcast #Fexingo #SpectrumCoordination

Keep every episode free: buymeacoffee.com/fexingo

Wind turbines suffer costly unplanned downtime, but a growing number of operators are moving predictive maintenance from the cloud to the edge. We walk through how Siemens Gamesa deployed edge nodes directly inside turbine nacelles to process vibration and temperature data on site, cutting latency from minutes to milliseconds and reducing false positives by 40 percent. We also explore the broader shift: why edge inference is becoming standard for remote industrial assets, the role of containerized models that update over low-bandwidth satellite links, and what the economics look like for a single 8-megawatt offshore turbine. Plus, a real-world failure case — a gearbox that was flagged six hours before it seized — and what that meant for maintenance scheduling. If you're tracking the industrial Internet of Things and the push toward real-time, this episode gives you one concrete architecture that's already in the field.

#EdgeComputing #PredictiveMaintenance #WindEnergy #SiemensGamesa #IndustrialIoT #RealTimeAnalytics #VibrationAnalysis #MachineLearning #Containerization #OffshoreWind #RenewableEnergy #TurbineMaintenance #EdgeInference #LowBandwidth #SatelliteConnectivity #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 89 of The Edge Computing Podcast with Fexingo zooms in on a specific use case: how edge computing enables autonomous inventory robots in massive warehouses. Lucas and Luna break down the decision by a major retailer to shift from cloud-dependent robots to on-board edge inference, reducing latency from 200 milliseconds to under 10 milliseconds. They walk through the hardware choices—NVIDIA Jetson modules, custom vision models—and the real-world impact: 35 percent faster cycle counts and 50 percent less network bandwidth. The conversation also touches on the cost trade-offs, the shift toward decentralized fleet management, and what this means for the next generation of warehouse automation. A tight, specific look at one problem and the edge solution that solved it.

#EdgeComputing #InventoryRobots #WarehouseAutomation #NVIDIAJetson #RealTimeAI #Latency #RetailTech #AutonomousRobots #DecentralizedFleet #ComputerVision #SmartWarehouse #LogisticsTech #CycleCounting #EdgeInference #Technology #BusinessPodcast #FexingoBusiness #TheEdgeComputingPodcast

Keep every episode free: buymeacoffee.com/fexingo