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

In this episode of The Edge Computing Podcast, Lucas and Luna explore how edge computing is transforming food safety testing from a days-long lab process into a real-time, in-factory detection system. They focus on the work of Clear Labs, a company using edge-based genomic sequencers and machine learning models to identify pathogens like Salmonella and E. coli on-site, cutting detection time from three days to under an hour. Lucas unpacks the technical architecture: the sequencer captures genetic data, an edge GPU runs a lightweight neural network trained on tens of thousands of pathogen genomes, and a local dashboard alerts quality-control teams instantly — with only summary data sent to the cloud. Luna brings in the cost angle: a single contamination recall can cost a company tens of millions, making the edge investment a clear ROI play. They also discuss how this approach addresses the limitations of traditional batch testing, including sampling gaps and lab backlogs. The episode closes with a look at how this model could scale to other food safety applications, like allergen detection and supply chain verification.

#EdgeComputing #FoodSafety #ClearLabs #RealTimeDetection #GenomicSequencing #MachineLearning #Salmonella #EColi #PathogenDetection #FoodIndustry #QualityControl #DistributedInfrastructure #LocalCompute #ROI #SupplyChain #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 117 of The Edge Computing Podcast examines how edge computing enables real-time load balancing and fault detection in electric vehicle charging networks. Lucas and Luna discuss a specific case: a regional charging network in California that deployed edge nodes at 50 charging stations to cut latency from 200 milliseconds to under 10 milliseconds. They explore the technical architecture, the role of local compute in dynamic pricing, and why the edge was chosen over cloud-only for reliability. The hosts also touch on the broader implications for grid stability as EV adoption accelerates. The episode is grounded in a real deployment, giving listeners a concrete understanding of how edge infrastructure solves a critical bottleneck in the EV ecosystem.

#EdgeComputing #Technology #ElectricVehicles #EVCharging #RealTimeComputing #LoadBalancing #GridStability #DynamicPricing #California #ChargingInfrastructure #Latency #LocalCompute #FexingoBusiness #BusinessPodcast #EdgeInfrastructure #DistributedComputing #IoT #SmartGrid

Keep every episode free: buymeacoffee.com/fexingo

Episode 116 of The Edge Computing Podcast with Fexingo explores how edge computing is transforming digital advertising by enabling real-time, privacy-compliant personalization. Lucas and Luna dive into the specific case of a major ad-tech platform that shifted from cloud-based bid processing to edge-based inference, cutting latency from 200 milliseconds to under 20. They discuss how device-side machine learning models process user behavior locally—without sending raw data to central servers—while still syncing anonymized aggregates for campaign optimization. The episode covers the technical architecture: lightweight neural networks running on smartphone GPUs, CDN-hosted model updates, and the trade-offs between personalization depth and battery life. They also examine the business implications: higher click-through rates, lower data egress costs, and compliance with evolving privacy regulations like GDPR and CCPA. No hype, just the real engineering and economics behind a technology that may change how hundreds of billions in ad spend are allocated.

#EdgeComputing #DigitalAdvertising #RealTimePersonalization #PrivacyByDesign #MachineLearning #CDN #LatencyReduction #AdTech #GDPR #CCPA #OnDeviceAI #ModelInference #ClickThroughRate #DataEgress #CampaignOptimization #FexingoBusiness #BusinessPodcast #Technology

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna dive into how edge computing is transforming drone delivery logistics by enabling real-time obstacle avoidance, route recalculations, and package handoffs without cloud round trips. They focus on the case of Zipline, the medical drone delivery company that operates hundreds of daily flights in Rwanda, Ghana, and now parts of the US. Lucas explains how each drone carries an onboard edge processor that runs computer vision models locally, cutting decision latency from hundreds of milliseconds to under 10 milliseconds. Luna raises questions about battery life trade-offs and edge server placement at launch sites. The hosts also touch on how edge computing differs from traditional cloud-dependent drone operations, why latency spikes during cellular handoffs can crash a drone, and what this means for the broader logistics industry. A practical, example-driven look at an invisible infrastructure layer that makes autonomous flight viable at scale.

#EdgeComputing #DroneDelivery #Zipline #LogisticsTech #Latency #ComputerVision #AutonomousFlight #RealTimeProcessing #CloudVsEdge #Rwanda #Ghana #MedicalDelivery #DroneNavigation #OnboardAI #Technology #FexingoBusiness #BusinessPodcast #EdgeInfrastructure

Keep every episode free: buymeacoffee.com/fexingo

Imagine a stadium with 60,000 fans where every seat hears perfect audio, regardless of where the performer stands. This episode unpacks how L-Acoustics, a French audio company, uses edge computing nodes placed under seats and in lighting rigs to process 60,000 audio channels per venue in real time. Lucas and Luna explain the specific latency challenge—under five milliseconds for a live mic—and why cloud processing fails. The episode walks through the signal chain: from the performer's wireless mic to a local edge node doing convolution reverb and EQ, then to the speaker arrays. They discuss how L-Acoustics' L-ISA system uses multiple edge servers to handle spatial audio for shows like Taylor Swift's Eras Tour and Cirque du Soleil. The hosts also explore the broader implication: any live event with hundreds of microphones and speakers becomes a distributed computing problem. If you've ever heard terrible stadium sound, this episode explains why that's finally changing.

#EdgeComputing #LiveAudio #LAcoustics #SpatialAudio #RealTimeProcessing #ConcertTechnology #TaylorSwift #ErasTour #CirqueDuSoleil #LISA #LowLatency #DistributedComputing #AudioEngineering #LiveSound #Technology #FexingoBusiness #BusinessPodcast #LocalCompute

Keep every episode free: buymeacoffee.com/fexingo

Pharmaceutical counterfeiting is a $200 billion global problem, and edge computing is becoming a critical line of defense. In this episode, Lucas and Luna explore how distributed inference at the point of dispensing—pharmacy counters, hospital supply rooms, and clinic storage—enables real-time verification of drug authenticity without relying on slow, centralized databases. They examine a pilot program by a major pharmacy chain that uses a handheld edge device to analyze packaging holograms and QR codes in under 200 milliseconds, catching fakes before they reach patients. The hosts discuss the technical architecture: lightweight machine learning models running on a custom ASIC, offline-first verification with blockchain-anchored hashes, and the trade-offs in accuracy versus speed. They also touch on the regulatory implications and why this application of edge computing could save thousands of lives annually in regions with weak supply chain oversight.

#EdgeComputing #PharmaceuticalSupplyChain #CounterfeitDrugs #RealTimeVerification #MachineLearning #Blockchain #ASIC #OfflineFirst #DrugSafety #SupplyChainSecurity #PharmacyTech #HealthTech #FexingoBusiness #BusinessPodcast #Technology #DistributedComputing #PatientSafety #AntiCounterfeiting

Keep every episode free: buymeacoffee.com/fexingo

Episode 112 of The Edge Computing Podcast unpacks a tension rarely discussed: 5G promised single-digit millisecond latency, but the core network architecture hasn't kept up. Lucas and Luna drill into a real-world bottleneck — backhaul from the cell tower to the regional data center. In metro areas, that round trip can add 20 to 30 milliseconds, negating the radio-level gain. The fix? Distributed edge nodes at the tower base or aggregation point. We walk through a specific deployment by a Tier 1 carrier in Chicago using off-the-shelf servers running Kubernetes to handle low-latency video analytics for a smart-traffic pilot. The node processes license-plate data locally, only sending aggregated metadata upstream. The result: end-to-end latency dropped from 34 milliseconds to 8 milliseconds. We discuss the economics — the carrier spent roughly $40,000 per node versus $2 million for a mini data center — and why the telecom industry is still resistant to this architecture. No hype, just the numbers and a clear explanation of why edge computing is the missing layer for 5G to deliver on its latency promise.

#EdgeComputing #5G #Latency #Telco #Kubernetes #SmartTraffic #Chicago #VideoAnalytics #Backhaul #DistributedInfrastructure #Telecom #CloudNative #NetworkArchitecture #Technology #FexingoBusiness #BusinessPodcast #EdgeNodes #RealTimeProcessing

Keep every episode free: buymeacoffee.com/fexingo

In this episode of The Edge Computing Podcast, hosts Lucas and Luna explore how edge computing is revolutionizing building energy optimization. They dive into the case of the Empire State Building, which reduced its energy consumption by 38% using a distributed network of edge nodes that process sensor data locally—no cloud latency. The conversation covers the technical architecture, real-time HVAC adjustments, and the business case for retrofitting older buildings. Lucas explains how local compute eliminates the round-trip to the cloud, enabling sub-second responses to changing conditions. Luna challenges the scalability for smaller buildings, and they discuss the role of open standards like BACnet. The episode also touches on the broader implications for urban sustainability, including a funding plug for the show. Specific numbers and real-world examples make this a practical guide for property managers and tech enthusiasts alike.

#EdgeComputing #BuildingEnergy #SmartBuildings #EmpireStateBuilding #HVAC #RealTimeAnalytics #DistributedInfrastructure #Sustainability #TechPodcast #EnergyOptimization #BACnet #IoT #SensorNetworks #PropertyTech #ClimateTech #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 110 of The Edge Computing Podcast explores how edge computing is transforming precision fermentation — the process of using engineered microbes to produce proteins, fats, and other compounds. Lucas and Luna dive into a specific case study: a startup called New Culture Foods in Berkeley, California, which uses edge nodes to monitor and control fermentation tanks in real time. They discuss the latency and data challenges that cloud-only architectures can't solve, the role of local inference models for pH and temperature adjustments, and how this technology is enabling a new wave of sustainable food production. The episode also touches on the broader implications for biomanufacturing, from alternative proteins to pharmaceutical ingredients, and why edge computing is becoming essential for any process that requires sub-second responses to biological signals.

#EdgeComputing #PrecisionFermentation #NewCultureFoods #AlternativeProteins #Biomanufacturing #RealTimeMonitoring #LocalCompute #FermentationTech #SustainableFood #BerkeleyStartups #IndustrialIoT #MachineLearningAtTheEdge #Bioengineering #LabGrownDairy #FoodTech #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

In episode 109 of The Edge Computing Podcast, Lucas and Luna explore how edge computing is making mixed reality (MR) collaboration viable for remote teams. They dive into a specific case: a global architecture firm using edge nodes to render shared holographic building models with sub-20-millisecond latency, eliminating the nausea and lag that plagued earlier cloud-dependent MR. Lucas explains why traditional cloud round-trips add 80–120 ms—too slow for seamless interaction—and how edge nodes at the firm's office and partner locations cut that to under 10 ms locally, with sync latency under 20 ms across offices. Luna highlights the role of 5G network slicing in prioritizing MR data packets. They touch on the economic trade-off: edge hardware costs versus productivity gains, and why this use case is a 'killer app' for distributed edge infrastructure. The episode includes a brief, sincere pitch supporting Fexingo's ad-free model via buymeacoffee.com/fexingo.

#EdgeComputing #MixedReality #SpatialComputing #RemoteCollaboration #5G #NetworkSlicing #Architecture #HolographicRendering #Latency #Sub20ms #CloudVsEdge #EnterpriseTech #RealTime #DigitalTwin #DistributedInfrastructure #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo