About
At Rectitude 369, my focus as Principal Architect is on orchestrating innovative network…
Articles by Christopher
-
🚀 AI ISN'T JUST FOR SOFTWARE DEVS ANYMORE — IT'S REVOLUTIONIZING INFRASTRUCTURE 🚀
🚀 AI ISN'T JUST FOR SOFTWARE DEVS ANYMORE — IT'S REVOLUTIONIZING INFRASTRUCTURE 🚀
Hot take: If you think AI-assisted coding is only for building apps, you're missing the BIGGEST infrastructure…
-
Technology Partnerships in Large Public Venues: Our Approach
Technology Partnerships in Large Public Venues: Our Approach
Transforming Challenges into Opportunities Rectitude 369 is leading the way in technology partnerships with LPVs…
Activity
1K followers
-
Christopher Nelson shared thisGood post, and one of my mantras for years (and have used Portainer for years in many different deployments/use cases since it was Beta). My core was/is virtualization, and grew up on Terminal Services and VMware (along with HyperV, Xen, Proxmox, and currently Morpheus) - over 30 years of it (funny enough, was the "first" production AllScripts EMR virtualized, if that tells you how long). Container adoption, even with my evangelism, has been tougher than it should be, and has been exactly why you state - the management/operation ecosphere. The complication & sprawl of Kubernetes hasn't helped for large deployment adoption. With the likes of Proxmox and Morpheus (currently, with an integrated KVM standalone, Swarm, or Kubernetes cluster built within first) - the native deploy, use, & operations of containerized workloads from the primary UI help drive greater adoption rates in general for sure - especially since Broadcom ruined VMware (like they tend to do). The mass exodus to these alternative stack providers is bringing more to the trenches. Anyone getting started in containers, Portainer is definitely my recommendation to anyone.Christopher Nelson shared thisIf you’ve spent most of your career in VMware-based environments, this will probably sound familiar. I wrote down some honest thoughts on why I avoided containers for years, what eventually changed, and why I think the real challenge isn’t technical at all. #EnterpriseIT #InfrastructureArchitectureI Used to Avoid Containers. Here’s Why I Don’t Anymore.I Used to Avoid Containers. Here’s Why I Don’t Anymore.
-
Christopher Nelson shared thisHuge from our partners at Ruckus!Christopher Nelson shared thisWe’re accelerating the shift to AV-over-IP 🚀 We're proud to announce the expansion of our Pro AV ICX® switch portfolio, specifically engineered to support the global transition from legacy video transport to Ethernet-based systems. These switches aren't just powerful; they’re AV-optimized and pre-configured out-of-the-box. Faster deployments, effortless scaling, and massive reduction in configuration resources for our customers. We are also making moves in the ecosystem: 🤝 Strategic Partnership: With Crestron Electronics, we’re ensuring full interoperability with the DM NVX® platform 🌐 New Alliance: We’ve joined the SDVoE Alliance to help standardize AV transport over Ethernet Whether managed via RUCKUS One®, SmartZone™, or as a standalone, we’re making AV networking simpler than ever. Read the announcement: https://lnkd.in/entwgQv9 #ProAV #AVoverIP #Crestron #SDVoE
-
Christopher Nelson shared thisExcellent simple graphic for datacenter power paths/types for those of you who are "electrically challenged" (such as myself). 😀
-
Christopher Nelson posted thisThe era of the "mega-app" is dying. And honestly? Good riddance. Here's what not many people are talking about: System Engineers are quietly becoming the new software developers. Not through bootcamps. Not through CS degrees. Through necessity. Over the last couple years, I've built ~150 purpose-driven tools & apps. Not products. Not startups. Just... solutions. That gap in your monitoring stack? 200 lines of Python. That annoying manual process eating 30 minutes of your day? A quick utility. That "feature request" sitting in some vendor's backlog for 3 years? Built it myself yesterday (no kidding). This is vibe coding. And it's fundamentally changing who builds software. The old model: → Identify problem → Search for enterprise solution → 6-month procurement cycle → $50k+ licensing → 18-month implementation → It still doesn't do the ONE thing you actually needed The new model: → Identify problem → Build exactly what you need → Ship it to yourself by lunch → Iterate when requirements change (because they always do) I'm not competing with Datadog or ServiceNow. I don't need to. I just need something that solves MY specific workflow, in MY environment, TODAY. The tools have caught up to the vision. AI-assisted development isn't replacing engineers - it's amplifying those who actually understand the problems. Domain expertise + vibe coding = dangerous efficiency. Some just say "dangerous" — citing security holes, spaghetti code, or technical debt. But that's where domain expertise comes in. When you actually understand the problem space, your choice of tools and prompts becomes surgical rather than reckless. You're not blindly accepting outputs; you're steering them with intent. Let's be real: the loudest critics tend to be traditional devs or companies feeling the disruption. I get it - watching someone with zero formal training spin up in a weekend what took your team a quarter is... uncomfortable. But instead of dismissing it, take inspiration. Adapt. Make your solutions better. The tools don't care about credentials; they care about clarity of thought and domain knowledge. The "danger" isn't in the approach - it's in using it without the expertise to validate results. The best part? These micro-tools compose. They integrate. They talk to each other. Before you know it, you've built a custom ecosystem that fits your workflow like a glove. Large-scale dev projects aren't going away. But the monopoly on "who gets to build software" is officially over. If you're a sysadmin, neteng, or ops person who hasn't started building your own tools yet—what are you waiting for? The barrier to entry just hit the floor. What's the most useful tool/app you've built for yourself? Drop it below - I'm always hunting for inspiration to add to my 1000 other ideas - my Github repos are stocked full of my own inspirations to share. 😄 #Vibe #Development #SysOps #DevOps #NetOps #AI
-
Christopher Nelson shared thisExcellent resource about operating AI/LLM services on Kubernetes. Highly recommended read for any datacenter-focused #DevOps, #NetEng, and/or #SysEng either thinking about, or starting out deploying private AI services within a Kubernetes stack. At #Rectitude369, we're in the midst of our infrastructure upgrades, mesh design, & POC to bring our private AI Cloud in-house across 5 of our U.S. regional datacenters—for our internal use, and future customer solutions delivering secure, private, & performant AI inferencing services alongside our many other offerings. Generally, knowledge of AI servicing can feel like drinking from a fire hydrant, but this read is a great resource to cut through the noise. It covers the full operational lifecycle—from model deployment with vLLM and KServe, to GPU scheduling, observability, autoscaling, and even advanced topics like disaggregated serving and LLM-aware routing. Key takeaways that resonated with our current work: 📦 Model packaging & registry strategies (OCI images, Modelcars, Safetensors vs GGUF) ⚡ GPU resource management & multi-GPU inference patterns (tensor vs pipeline parallelism) 📊 LLM-specific observability metrics (TTFT, ITL, KV cache utilization) 🔄 Production tuning—quantization, runtime optimization, and intelligent request routing Whether you're evaluating self-hosted LLM infrastructure or scaling existing deployments, this read is a solid foundation to start with. #Kubernetes #GenerativeAI #LLM #MLOps #InfrastructureAsCode #AI #EdgeComputing #DataCenter #vLLM #LPV #NoCRavenChristopher Nelson shared thisExplore a framework for scaling gen AI workloads using Kubernetes and cloud-native methodologies to achieve high performance and cost efficiency.
-
Christopher Nelson shared thisThe ONLY NCM solution I trust! After 30+ years in IT (LPVs, Healthcare), rConfig is the most comprehensive NCM available. Most tools do basic backups. rConfig excels by backing up virtually ANY network-connected device via SSH, Telnet, SNMP, or crucially, API access. Throw in distributed automation & this is unmatched. For complex environments (LPVs, hospitals), rConfig covers BAS/HVAC, lighting, security, A/V (Daktronics, Dante, Crestron), and more. It delivers functionality, support, & compliance in mission-critical settings. Want to know more? Reach out! #Rectitude369 #rConfig #NCM #Networking #ITInfrastructure #LPV #BAS #AutomationChristopher Nelson shared this🚀 New Connection Template Drop — Just in Time for rConfig v8 Core We’ve just pushed a brand new connection template for Siemens SIEMENS-RUGGEDCOM devices into the public rConfig repo 👀 If you’re running @SIEMENS-RUGGEDCOM in industrial, utility, or harsh environments, this one’s for you. The template makes onboarding cleaner, faster, and way less painful — exactly how it should be. And the timing? Not accidental. This lands days before the release of rConfig v8 Core 💥 v8 Core is all about: Open, community-driven config management Better device onboarding out of the box Real-world vendor coverage (not just the usual suspects) This new RUGGEDCOM template is a great example of where v8 Core is heading — practical, production-ready, and built with real networks in mind. 👉 Learn more about rConfig v8 Core here: https://lnkd.in/d2DFccsb More templates. More vendors. More momentum. v8 Core is nearly here.rConfig V8 Core | Free Open-Source Network ConfigurationrConfig V8 Core | Free Open-Source Network Configuration
-
Christopher Nelson shared thisDefinitely a gamechanger!Christopher Nelson shared thisNVIDIA Vera Rubin is in full production. NVIDIA Vera Rubin NVL72 is a unified, rack-scale AI system that brings together 72 #NVIDIARubin GPUs and 36 NVIDIA Vera CPUs, designed to operate as a single, massive compute engine. Built to deliver unprecedented efficiency and performance for training, inference, and agentic AI at scale. #CES2026
-
Christopher Nelson shared this🛠️ If you think AI-assisted coding is only for building apps, you're missing the BIGGEST infrastructure revolution of the decade.🚀 AI ISN'T JUST FOR SOFTWARE DEVS ANYMORE — IT'S REVOLUTIONIZING INFRASTRUCTURE 🚀🚀 AI ISN'T JUST FOR SOFTWARE DEVS ANYMORE — IT'S REVOLUTIONIZING INFRASTRUCTURE 🚀
-
Christopher Nelson shared thisReally digging OpenCode. Having used Warp Terminal since Alpha the last couple years, and since becoming a micro-transactional rabbit-hole - we have ventured around at alternatives to use across the organization. I have been using various products the last couple months and OpenCode is one of the top contenders thus far (CLI and UI) - and really digging it. The CLI "UI" is the beautiful and preferred over other solutions (Warp, KiloCode, Goose/Code, Gemini, Co-Pilot, etc.). The ability to utilize your own local LLM is great (and a requirement with us). Highly recommended! 👍Christopher Nelson shared thisOpenCode, now on your desktop (in beta) https://lnkd.in/gHdebP3q
Christopher Nelson replied to a comment
2mo
John Xie hi John - absolutely. I will send you an email shortly, regarding something that came up internally here regarding Taskade just yesterday.
Christopher Nelson commented on a post
10mo
Grats goes out to the OKC Thunder on their 1st championship. We are proud partners, and looking forward to the future.
Broke some data records across the network/WiFi tonight, was awesome to see and be a part of.
#ThunderUp #OKCProud #Rectitude369 #LPV #Elite #PaycomCenter #ASMGlobal #ASMOKC #WiFi
-
Christopher Nelson liked thisChristopher Nelson liked thisWe are #hiring! Know anyone who might be interested?
-
Christopher Nelson liked thisChristopher Nelson liked thisIf you’ve spent most of your career in VMware-based environments, this will probably sound familiar. I wrote down some honest thoughts on why I avoided containers for years, what eventually changed, and why I think the real challenge isn’t technical at all. #EnterpriseIT #InfrastructureArchitectureI Used to Avoid Containers. Here’s Why I Don’t Anymore.I Used to Avoid Containers. Here’s Why I Don’t Anymore.
-
Christopher Nelson liked thisChristopher Nelson liked thisThe scale of AI Infrastructure is wild. And we're just getting started. My good buddy Keith ONeill described our current situation as "trying to build cloud computing with dial up modems." But the modems are 800Gbps interfaces! We literally can't throw enough bandwidth at AI Infrastructure right now. We need faster chips, bigger interfaces, and higher capacity carriers. On that last, one story I heard over and over again from Pacific Telecommunications Council (PTC) last month was the difficult hunt for more transit. This week at NANOG we heard from Meta's Omar Baldonado about their AI data centers and I just love the rack porn... This is what AI Infrastructure actually looks like. What Each Rack Represents: MI300X (AMD): AMD Instinct with 192GB HBM3 memory. Used for Llama 3/4 inference, recommendations, and ads workloads. MTIA: Meta's custom silicon (5nm RISC-V). Optimized for ranking/recommendation models. Deployed in 16 data centers, 3x faster than v1, 90W per chip. H200 (NVIDIA Hopper): 141GB HBM3e memory, 4.8TB/s bandwidth. Used for training and inference requiring high memory capacity. GB200 (NVIDIA Blackwell): Meta's "Catalina" rack. 36 Grace CPUs + 72 B200 GPUs, ~140kW, liquid-cooled. 360 PFLOPS, 30x faster inference vs H100. GB300 (Blackwell Ultra): 72 B300 GPUs, 288GB/GPU, 1.1 exaflops FP4 compute, 800Gbps networking. Meta's most advanced AI infrastructure. -- #aiinfrastructure #artificialintelligence #networkinfrastructure #nanog96
-
Christopher Nelson liked thisChristopher Nelson liked thisHey everyone! 👋 Interactive maps are now supported in your apps! 📍🗺️ 🏠 Real estate portals with property maps 📍 Service directories searchable by location 🎪 Event finders with venue maps 🗺️ Location trackers for teams/services Prompt example: "Build a restaurant finder app with an interactive map." 🧬 Try it: https://lnkd.in/gpCcd2vX ✨ Share your map-ready apps to get featured!
-
Christopher Nelson liked thisChristopher Nelson liked thisAgentic memory framework for LLMs and AI Agents! MemU is an open-source agent memory framework that lets LLMs store, organize, and reason over long-term memory using a file-system based design. Instead of stuffing context or relying only on vector search, MemU lets agents read and reason over memory files directly. Memory is not an index. It’s something the model can understand. MemU ingests multimodal inputs, extracts structured textual memory items, and autonomously organizes them into thematic Markdown files. How memory is structured: Raw resources → memory items → memory category files Documents, conversations, images, and audio are preserved in their original form, without deletion or modification. Facts are then extracted and organized into human-readable memory category files. Key features: • Dual-mode retrieval, including LLM-based (non-embedding) search for higher accuracy • File-system based memory where each category is a Markdown file • Hierarchical memory layers that preserve traceability • Native multimodal memory for text, images, audio, and video • Lightweight and developer-friendly, no heavy graph constraints • Fully configurable prompts for high extensibility Why this architecture matters: Most memory systems force developers to decide what matters. MemU lets the agent decide. It learns what to remember, promotes frequently used knowledge, and reorganizes memory as usage evolves. Retrieval works top-down and falls back gracefully when needed. The result is better temporal reasoning, fewer hallucinations, and memory that actually scales across sessions. The best part? It’s 100% open source. Link to the GitHub repo in the comments!
-
Christopher Nelson liked thisGet ProAV capabilities that work!Christopher Nelson liked thisWe’re accelerating the shift to AV-over-IP 🚀 We're proud to announce the expansion of our Pro AV ICX® switch portfolio, specifically engineered to support the global transition from legacy video transport to Ethernet-based systems. These switches aren't just powerful; they’re AV-optimized and pre-configured out-of-the-box. Faster deployments, effortless scaling, and massive reduction in configuration resources for our customers. We are also making moves in the ecosystem: 🤝 Strategic Partnership: With Crestron Electronics, we’re ensuring full interoperability with the DM NVX® platform 🌐 New Alliance: We’ve joined the SDVoE Alliance to help standardize AV transport over Ethernet Whether managed via RUCKUS One®, SmartZone™, or as a standalone, we’re making AV networking simpler than ever. Read the announcement: https://lnkd.in/entwgQv9 #ProAV #AVoverIP #Crestron #SDVoE
Experience & Education
-
Rectitude 369, LLC
View Christopher’s full experience
See their title, tenure and more.
or
Licenses & Certifications
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content