Michael Hoffmann’s Post

1/ As we build out Corbits, an internal metric we use to measure progress is how easy our tooling is to use. We do a lot of dogfooding. This last week we brought @0xPolygon and Polygon Amoy online for x402 payments using Corbits and Faremeter. 2/ As more chains show interest in x402, it is important they can come online without needing to wait for PRs or needing to fork software. Working with @0xPolygon we're happy to share that all EVM builders can now build x402 apps using our OSS x402 framework @faremeterxyz 3/ Using Faremeter we were able to add support for Polygon in under 10 minutes from start to finish. And with full compatibility across all Faremeter plugins. Now, if you're building on Polygon and interested in x402 you can use our public facilitator: facilitator.corbits.dev/ 4/ At Corbits we want to x402 everything. That means unlocking the best tools for x402 on any chain. If you're working to bring x402 to your network and are running into blockers, our DMs are always open. We build with you. https://lnkd.in/e3rrhjnz

"This Helm chart is the pragmatic answer. It’s still three separate backends under the hood (Loki, Tempo, Mimir), but thanks to OpenTelemetry’s OTLP protocol, it feels like one unified system. Your apps emit OTLP, hit a single gateway, and that gateway routes everything to the right storage. You get centralized ingestion, Grafana pre-wired for correlation, and zero vendor lock-in." https://lnkd.in/eaBGHds7

Distributed tracing with OpenTelemetry changed our approach to debugging Go microservices. Before: grep through logs of 27 services, manual timeline reconstruction, guessing games. After: one trace ID → visualization of entire request flow → instant problem identification. Payment Service slowing down to 2.8s? Visible immediately. N+1 query after deployment? Found in 15 minutes. Connection leak? Discovered faster than water boils. Full implementation guide with Go examples: https://lnkd.in/dnYdFKqg #golang #microservices #observability #opentelemetry #tempo #jaeger #distributedtracing #sre

Tooling update: edge runtimes like TensorFlow Lite Micro, ONNX Runtime, and Arm NN are evolving fast. Which runtimes do you trust for production edge apps and why? #EdgeAI #TinyML

Today we're releasing Cap v0.3.73, which includes a new window excluder feature. Here's a live example of adding ChatGPT to the list of excluded windows in Cap, and then recording via Studio Mode with ChatGPT still in the frame. Full changelog below 👇

🚀 NetsLab Demonstration Video Series – Secure Open RAN with Hierarchical Authentication and Authorization 🔐📡 In this video, we demonstrate a practical implementation of our hierarchical security framework for Open RAN as proposed in the paper “Securing xApps in Open RAN: A Hierarchical Approach to Authentication and Authorisation” (IEEE CNS 2025 - 📚 Read the full paper on IEEE Xplore: https://lnkd.in/gN_7pvzi). The demo highlights how the proposed framework strengthens the authentication and authorisation of third-party xApps and rApps within RAN Intelligent Controllers (RICs) while maintaining scalability and minimal performance overhead. We showcase two experimental setups that simulate non-RT RIC and near-RT RIC environments running on virtualised machines. Using an enhanced XRF (OAuth 2.0) server, the demo illustrates the flow of access tokens, refresh tokens, and token rotation, providing fine-grained access control and protection against token leakage. 🎥 Watch the Demo Video 🔗 https://lnkd.in/gT3ifWsN In the demo, we walk through: • Deployment of client and resource xApps communicating via secured APIs • Hierarchical token-based authorisation and inter-RIC coordination • Token rotation and revocation mechanism to detect and mitigate token hijacking • Comparative results on authentication delay, refresh latency, and bandwidth efficiency The proposed framework delivers a scalable and secure solution for third-party application onboarding in Open RAN, reducing the attack surface while preserving system performance. 🔑 Key Highlights • Hierarchical security model for rApps, xApps, and dApps • Improved XRF framework with OAuth 2.0 access + refresh JWT tokens • Token rotation mechanism for leak detection and revocation • Lightweight overhead and fast authentication under multi-threaded loads #OpenRAN #RIC #xApps #rApps #5GSecurity #6GResearch #OAI #NetworkSecurity #FutureNetworks #IEEE #NetsLab Pramitha Fernando An Braeken Pawani Porambage Madhusanka Liyanage Vrije Universiteit Brussel | VTT | University College Dublin | UCD School of Computer Science CONNECT Centre Research Ireland UCD Research ENSURE-6G CONFIDENTIAL6G Project European Commission 6G Smart Networks and Services Industry Association

Check out the latest NetsLab demo, which presents our Secure Open RAN framework with Hierarchical Authentication and Authorisation, built upon the paper that received the Best Paper Award at IEEE CNS 2025 🏆. This demonstration highlights a scalable and secure approach for third-party xApp and rApp authentication and authorization in Open RAN environments. #OpenRAN #6G #NetworkSecurity #IEEE #NetsLab

VoiceQ 9.0's AI reduces lip sync word timing to original media or imported audio by up to 75%. Precision adjustments and intelligent video analysis make the difference. Available for macOS. #LipSync #VoiceQAI #ADRtools

Just published this #OpenSource library to make your LLM apps or AI agents more reliable. All you need is a few lines of code to call any LLM service API with automatic rate limits (token bucket rate algo), smart retries, backoff, and circuit breaker. Source Code - https://lnkd.in/gCMUHCgE Cheers to the spirit of #Hacktoberfest

Real time systems need fast communication. But not all fast communication is created equal. I’ve seen teams struggle to choose between gRPC and WebSocket especially when building chat systems, live data streams, or reactive backends. Here’s how I think about the difference: Use gRPC when: – You need structured, strongly typed requests and responses – You want clear service definitions and version control – You’re building communication between backend services – You need features like built-in retries, timeouts, and authentication Use WebSocket when: – You need low latency, two way streaming – You are building highly interactive user interfaces like chat apps, games, or live dashboards – You control both the client and the server – You want a long lived connection that allows the server to push updates to the client In one of our projects, we used gRPC for communication between internal services. And we used WebSocket to stream real-time updates to a frontend dashboard. Same platform. Two protocols. Two different use cases. There’s no such thing as “better” only what fits your architecture. How do you decide between them in your systems? #SystemDesign #Microservices #gRPC #WebSocket #RealTimeArchitecture #BackendEngineering #SpringBoot #CloudEngineering #SoftwareArchitecture