Essays, engineering notes, and open-source stories by Arun Babu Neelicattu.https://abn.is/en-ushttps://abn.is/void/scaling-dbos-from-prototyping-to-production-grade-durability/https://abn.is/void/scaling-dbos-from-prototyping-to-production-grade-durability/When you first start with DBOS, the magic is in the simplicity: your database is the orchestrator. You don't need to manage a separate cluster or navigate complex RPC configurations—you just use Postgres. 🙋If you are here and are wondering what DBOS is, then maybe check out their docs here. But as your application matures and your workflow throughput hits production-level scale, that same simplicity can become your biggest bottleneck if not handled correctly. Scaling DBOS isn't about throwingWed, 11 Mar 2026 20:46:55 GMTPostgreshttps://abn.is/void/evolving-our-collaborative-product-development-mantra/https://abn.is/void/evolving-our-collaborative-product-development-mantra/The "What vs. How" split often builds silos instead of bridges. By reframing our work as advocacies for the Why, the What, and the How, we can move beyond the "feature factory" and build products that deliver meaningful value to our customers and purpose to our teams.Mon, 23 Feb 2026 19:09:29 GMThttps://abn.is/void/wifi-on-ice-systemd-resolved/https://abn.is/void/wifi-on-ice-systemd-resolved/Struggling with WIFIonICE on your Linux distro? Your secure DNS configuration might be blocking the login page. This guide provides a quick fix to get you online on Deutsche Bahn trains without compromising your privacy on other networks.Wed, 08 Oct 2025 20:49:09 GMTHow Tohttps://abn.is/void/leadership-debt/https://abn.is/void/leadership-debt/Indecisive leadership creates "Leadership Debt," a hidden liability that stalls projects, demotivates teams, and cripples innovation. Building psychological safety and contextual decision-making systems is key to high-performing engineering teams.Fri, 29 Aug 2025 21:57:31 GMTLeadershiphttps://abn.is/void/beyond-the-buzzword-tapps/https://abn.is/void/beyond-the-buzzword-tapps/Radical Candor alone can backfire if trust and safety are missing. Build resilient engineering teams by focusing first on psychological safety and the TAPPs pillars: Trust, Autonomy, Purpose, and Psychological Safety. Move beyond buzzwords—become a cultural architect.Thu, 28 Aug 2025 00:28:51 GMTLeadershiphttps://abn.is/void/the-fabric-schism-is-over-why-your-next-move-should-be-towards-open-ethernet/https://abn.is/void/the-fabric-schism-is-over-why-your-next-move-should-be-towards-open-ethernet/The painful choice between proprietary performance and open economics is over. The fabric schism has ended. As performance equalizes, the overwhelming TCO and strategic advantages of open Ethernet make it the only logical path forward for AI at scale.Tue, 26 Aug 2025 00:00:43 GMTNetworkhttps://abn.is/void/blueprint-for-an-ai-supercluster-a-reference-architecture-for-100-000-accelerators/https://abn.is/void/blueprint-for-an-ai-supercluster-a-reference-architecture-for-100-000-accelerators/How do you actually build a 100,000-GPU cluster? This blueprint outlines a practical reference architecture: a hybrid model using proprietary links for scale-up and open UEC Ethernet for the massive scale-out, all tied together by the DPU.Thu, 21 Aug 2025 01:27:08 GMTNetworkhttps://abn.is/void/how-to-use-kubernetes-registry-remapping-to-redirect-bitnami-images-to-bitnami-legacy-as-a-temporary-migration-mitigation/https://abn.is/void/how-to-use-kubernetes-registry-remapping-to-redirect-bitnami-images-to-bitnami-legacy-as-a-temporary-migration-mitigation/Bitnami is moving older container images from docker.io/bitnami to docker.io/bitnamilegacy in August 2025. Use Kubernetes containerd registry remapping via a DaemonSet to transparently redirect image pulls, ensuring smooth migration without immediate manifest changes.Mon, 18 Aug 2025 23:18:07 GMTKuberneteshttps://abn.is/void/performance-adjusted-tco-the-real-math-behind-choosing-your-ai-network/https://abn.is/void/performance-adjusted-tco-the-real-math-behind-choosing-your-ai-network/The most expensive network isn't always the best. As open Ethernet closes the performance gap, the real math comes down to Performance-Adjusted TCO. When speed is equal, the lower cost of hardware and people becomes the deciding factor.Mon, 18 Aug 2025 22:38:38 GMTNetworkhttps://abn.is/void/from-monolith-to-mosaic-how-composable-ai-is-redefining-enterprise-value/https://abn.is/void/from-monolith-to-mosaic-how-composable-ai-is-redefining-enterprise-value/Enterprise AI is shifting from giant, costly models to efficient, specialized Small Language Models (SLMs) working in teams. This composable approach boosts accuracy, cuts costs, enhances privacy, and enables on-premise deployment—transforming AI into a competitive advantage for 2026 and beyond.Fri, 15 Aug 2025 22:00:40 GMTArtificial Intelligencehttps://abn.is/void/the-vendor-wars-nvidias-integrated-fortress-vs-the-open-ecosystem-alliance/https://abn.is/void/the-vendor-wars-nvidias-integrated-fortress-vs-the-open-ecosystem-alliance/The battle for AI networking is a clash of philosophies. NVIDIA's integrated fortress offers turnkey performance at a proprietary cost. The Open Alliance bets on collaboration and open standards. Your choice is more than tech—it's a strategic vote.Thu, 14 Aug 2025 00:07:13 GMTNetworkhttps://abn.is/void/meet-the-dpu-the-third-pillar-of-compute-that-makes-ai-clouds-possible/https://abn.is/void/meet-the-dpu-the-third-pillar-of-compute-that-makes-ai-clouds-possible/Your multi-million dollar GPUs are wasting time on chores. The DPU is the third pillar of compute, a specialized processor that offloads infrastructure tasks, freeing up your CPU/GPU and enabling the entire business model of the secure AI cloud.Tue, 12 Aug 2025 01:58:54 GMTNetworkhttps://abn.is/void/openllmetry-vs-openinference-the-observability-stack-your-llm-app-needs/https://abn.is/void/openllmetry-vs-openinference-the-observability-stack-your-llm-app-needs/Your app looks fine on the surface, but users say it's failing. Traditional logs can't explain why. Enter OpenLLMetry and OpenInference—tools that bring clarity to AI observability by capturing rich, standardized telemetry from your LLM stack.Fri, 08 Aug 2025 22:41:41 GMTObservabilityhttps://abn.is/void/packet-spraying-flexible-order-the-uecs-secret-weapons-for-ai-networking/https://abn.is/void/packet-spraying-flexible-order-the-uecs-secret-weapons-for-ai-networking/Traditional networks create traffic jams. Ultra Ethernet's secret weapons are packet spraying and flexible order—a new design that eliminates hotspots and unlocks the full power of the fabric by breaking all the old rules of packet delivery.Wed, 06 Aug 2025 21:29:36 GMTNetworkhttps://abn.is/void/not-your-standard-ethernet-the-industry-wide-revolution-to-dethrone-infiniband/https://abn.is/void/not-your-standard-ethernet-the-industry-wide-revolution-to-dethrone-infiniband/Standard Ethernet can't handle AI. Early attempts were brittle patches on a flawed foundation. Now, a massive industry alliance is re-architecting it from the ground up. This is the story of the Ultra Ethernet Consortium's revolution.Mon, 04 Aug 2025 22:30:31 GMTNetworkhttps://abn.is/void/a-software-engineers-guide-to-llm-observability-and-evaluation/https://abn.is/void/a-software-engineers-guide-to-llm-observability-and-evaluation/LLM observability is key for reliable AI apps. Traditional monitoring misses semantic errors. This guide covers four essentials: tracing, evaluation, prompt management, and analytics to build robust generative AI solutions.Fri, 01 Aug 2025 21:34:56 GMTObservabilityhttps://abn.is/void/run-npm-packages-without-installing-node-js/https://abn.is/void/run-npm-packages-without-installing-node-js/Want to use an npm package without installing Node.js? Learn how to use a simple shell alias with Podman or Docker to run any tool in an isolated container. This keeps your host system clean while giving you on-demand access to any package. Here’s how to set it up.Thu, 31 Jul 2025 23:41:39 GMTHow Tohttps://abn.is/void/infiniband-the-king-of-ai-networking-and-the-walls-of-its-walled-garden/https://abn.is/void/infiniband-the-king-of-ai-networking-and-the-walls-of-its-walled-garden/For over two decades, InfiniBand has powered the fastest AI clusters through native lossless design, RDMA, and in-network computing—making it the performance king. Yet, NVIDIA’s tight integration and high costs have hyperscalers seeking open, Ethernet-based alternatives.Wed, 30 Jul 2025 23:56:52 GMTNetworkhttps://abn.is/void/why-your-ai-factorys-biggest-bottleneck-isnt-the-gpu-its-the-network/https://abn.is/void/why-your-ai-factorys-biggest-bottleneck-isnt-the-gpu-its-the-network/AI breakthroughs aren’t just about GPUs—networks are the real bottleneck. Large-scale AI training jobs depend on rapid, lossless communication; even a 0.1% packet loss can slash GPU utilization by 13%. Job completion time now depends on robust, predictable network design.Mon, 28 Jul 2025 21:49:17 GMTData Centerhttps://abn.is/void/the-silent-trust-killer-is-cognitive-dissonance-poisoning-your-workplace/https://abn.is/void/the-silent-trust-killer-is-cognitive-dissonance-poisoning-your-workplace/Is your workplace saying one thing but doing another? Discover how cognitive dissonance erodes trust, sparks stress, and what leaders can do to build a culture of true alignment and engagement.Fri, 25 Jul 2025 14:08:31 GMTLeadershiphttps://abn.is/void/the-unseen-revolution-why-small-models-could-be-the-true-engine-of-agentic-ai/https://abn.is/void/the-unseen-revolution-why-small-models-could-be-the-true-engine-of-agentic-ai/Agentic AI is transforming the field, enabling fast, efficient, and affordable automation far beyond chatbots through small, specialized models. These Small Language Models (SLMs) can outperform large models in real-world tasks, powering the next generation of enterprise and on-device AI.Mon, 21 Jul 2025 23:38:41 GMTArtificial Intelligencehttps://abn.is/void/hey-kubernetes-devs-youre-probably-an-openshift-pro-and-dont-even-know-it/https://abn.is/void/hey-kubernetes-devs-youre-probably-an-openshift-pro-and-dont-even-know-it/Know Kubernetes? You're closer to mastering OpenShift than you think. It's the K8s you know, but with enterprise power-ups built-in. Learn the key differences—from its strict security model to traffic management—and use our cheat sheet to get your apps running on OpenShift today.Sun, 20 Jul 2025 17:33:29 GMTOpenShifthttps://abn.is/void/choosing-the-right-communication-protocol-in-2025-grpc-traditional-http-and-the-evolving-landscape/https://abn.is/void/choosing-the-right-communication-protocol-in-2025-grpc-traditional-http-and-the-evolving-landscape/The post explores modern communication protocols—REST, GraphQL, gRPC, HTTP/2, HTTP/3, Protocol Buffers over HTTP, and WebTransport—offering practical recommendations for backend, frontend, and API layers, examining tradeoffs in performance, compatibility, and maintainability.Sat, 19 Jul 2025 23:41:22 GMThttps://abn.is/void/the-handbrake-on-innovation-why-your-it-security-is-costing-you-more-than-you-think/https://abn.is/void/the-handbrake-on-innovation-why-your-it-security-is-costing-you-more-than-you-think/Legacy IT security models stifle innovation, forcing engineers to fight slow VPNs and restrictive laptops. Zero Trust shifts focus: never trust, always verify. It removes friction, boosts productivity, and protects your business, unlocking true agility and competitive advantage.Sat, 19 Jul 2025 16:19:01 GMTSecurityhttps://abn.is/void/completing-kubectl-plugins/https://abn.is/void/completing-kubectl-plugins/Learn how to configure shell completion for your plugins and boost your productivity with easy auto-complete!Sat, 19 Jul 2025 01:09:16 GMTHow Tohttps://abn.is/void/a-python-project-postgresql-and-wasm/https://abn.is/void/a-python-project-postgresql-and-wasm/Tired of slow, complex database testing? Meet py-pglite: run real PostgreSQL in your Python tests, no containers or mocks needed. Fast, zero-config, and Postgres-native—testing just got easierSun, 29 Jun 2025 00:59:25 GMThttps://abn.is/void/the-cautionary-tale-of-innovatenow-how-we-staged-a-security-play-and-watched-the-set-burn-down/https://abn.is/void/the-cautionary-tale-of-innovatenow-how-we-staged-a-security-play-and-watched-the-set-burn-down/InnovateNow faked security to win a big client—policies no one read, passwords no one remembered, and a shiny compliance cert. It worked, until a simple phishing email exposed it all. The breach wasn't dramatic, just inevitable. Real security isn't theater. It's culture, enforcement, and honesty.Fri, 27 Jun 2025 22:23:26 GMTSecurityhttps://abn.is/void/a-pragmatists-guide-to-open-source-licensing-from-fear-to-innovation/https://abn.is/void/a-pragmatists-guide-to-open-source-licensing-from-fear-to-innovation/Every modern software product is a tapestry woven with threads of open source code. From the operating system it runs on to the libraries that power its features, OSS is the engine of innovation. Yet for many startups, product managers, and even seasoned engineers, the world of OSS licenses is a landscape of fear, uncertainty, and doubt (FUD). The most common fear? "If we use a single open source library, we'll be forced to open-source our entire proprietary codebase." Let's be clear: This is Sun, 22 Jun 2025 01:07:06 GMTOpen Sourcehttps://abn.is/void/automating-llms-txt-context-for-the-lazy-engineer/https://abn.is/void/automating-llms-txt-context-for-the-lazy-engineer/TL;DR Programmer was lazy to manually upload text files to AI coding agents. Found an MCP server that helps, but was too lazy to keep updating the command arguments. Spent a couple of hours writing a daemon that reloads server with required changes on config or directory changes.Wed, 18 Jun 2025 17:53:10 GMTModel Context Protocolhttps://abn.is/void/are-we-building-ai-agents-inside-out/https://abn.is/void/are-we-building-ai-agents-inside-out/In the race to build powerful AI systems, are we focused on the wrong thing? New research suggests success lies not in agent skill alone, but in the architecture of their prompts and collaboration. This piece explores how to think less like a trainer and more like a system architect.Tue, 10 Jun 2025 18:34:55 GMTArtificial Intelligencehttps://abn.is/void/discovering-and-multiplexing-mcp-servers-using-agent-gateway/https://abn.is/void/discovering-and-multiplexing-mcp-servers-using-agent-gateway/ ⚠️This article was based on Agent Gateway version 0.5.2. Releases after this version have introduced a different configuration schema. You can see new example configuration on the project repository. Recently I have been looking at how to deploy multiple model context protocol (MCP) servers in a Kubernetes environment such that agents can discover and use them dynamically as they are deployed. This led me to a project that was originally called MCP Proxy which in months of its inception aMon, 09 Jun 2025 18:43:06 GMTModel Context Protocolhttps://abn.is/void/deconstructing-digital-minds-a-blueprint-for-language-agents/https://abn.is/void/deconstructing-digital-minds-a-blueprint-for-language-agents/Large Language Models (LLMs) are rapidly changing how we interact with technology. From answering our questions to drafting emails, their capabilities are impressive. But as we push them to do more complex tasks – to act as "agents" that can reason, plan, and interact with the digital or even physical world – we've entered a bit of a Wild West. Many different approaches are emerging, each with its own terminology and design, making it hard to compare them or build systematically on previous workThu, 15 May 2025 21:49:11 GMTArtificial Intelligencehttps://abn.is/void/containing-local-mcp-servers/https://abn.is/void/containing-local-mcp-servers/It would seem that Model Context Protocol is the talk of the town these days. With the onset of tools like Anthropic's Claude Code and JetBrain's Junie being able to run local instances MCP Servers are helping to provide relevant context to AI Agents "just in time". The quality and reliability of MCP Servers available today is rather variable. And it will take some time, as with most software, for things to get better. Today, a lot of the MCP Servers out there are hacked together or vibe coded.Sun, 11 May 2025 17:17:14 GMTModel Context Protocolhttps://abn.is/void/ghost-lazy-loading-mermaid-rendering-script/https://abn.is/void/ghost-lazy-loading-mermaid-rendering-script/Recently, I started trying out Ghost for publishing posts. Having been an Obsidian user for some time now, I have learned to truly appreciate Mermaid for diagramming as Obsidian has out of the box support for it. Ghost on the other hand, does not support rendering of Mermaid code blocks without some additional effort. Researching on how to get it setup, I came across the Ghost Mermaid JS post by Samuel Hierholzer. The post details how to get it setup, and it works really well. The summary of thSun, 30 Mar 2025 00:08:43 GMThttps://abn.is/void/dbos-vs-temporal-choosing-your-durable-executor/https://abn.is/void/dbos-vs-temporal-choosing-your-durable-executor/Backend development increasingly demands applications that are not only scalable but also inherently reliable. Durable execution platforms like DBOS and Temporal have emerged as viable options for production software, offering solutions for building and operating complex, distributed systems. This post provides a comparison to help software engineers, particularly those targeting on-premise deployments, understand the trade-offs between these two powerful tools. In this post, we assume a foundaThu, 27 Mar 2025 19:36:27 GMTSoftware Engineering