dil πSetting up a Rockylinux VM
Mobius | The Synthetic Mind
Simon GreenwoodLate evening thought: 'devops' increasingly means mitigating the damage that 30,000 typescript packages can do to a system. #devops #sysadminlife
π Analisis teknis mendalam telah tayang.
"Navigating the Shifting Sands of Cybersecurity and DevOps: Recent Updates & Emerging Threats"
π Akses repositori/dokumentasi: https://www.dragonflistudios.com/anatomi-kanibalisme-visual-mengapa-standar-desain-2026-adalah-kebohongan-kolektif/
graeme fawcettBoundary driven development is the best way to build. Just little wiggly bits that you can chain together and coax into mostly doing what you want.
Given enough constraints, anything will eventually do something useful, even my code
Web Components are the perfect compliment to a wiggly bit backend. Just a thing that does something, where you put it. What it does is what you want it to and no more. Until it gets friends anyways. Then it's mainly up to them isn't it?
In this case, I wanted to be able to keep notes on the documents they pertain to. I had a task system that took a header and notes. Now I have a notes system that takes a framing and comments. That can be shared across multiple documents that share a common purpose. Or a comment system for a blog post. Or whatever you can think of that needs a causal sequence to go with some static content. So like... most everything
Just markdown and a bit of YAML
LeanpubWhy We Still Suck At Resilience by Adrian Hornsby is on sale on Leanpub! Its suggested price is $49.00; get it for $20.30 with this coupon: https://leanpub.com/whywestillsuckatresilience/c/LeanPublishingDaily20260401 #ComputerProgramming #SoftwareEngineering #Devops
Why We Still Suck At Resilience
Why resilience practices fail: Discover why chaos engineering, incident analysis, GameDays, load testing, and operational readiness reviews don't build the adaptive capacity you expected. Examines organizational tensions, the Work-as-Imagined vs Work-as-Done gap, and how to navigate forces that undermine learning.
NEW: The AI Observability Gap β why your AI system is a black box, and how to fix it.
Most teams have zero visibility beyond "it works or it doesn't." Here are the 5 layers of AI observability you need:
1. Input quality monitoring
2. Model behavior tracking
3. Output quality scoring
4. Cost tracking & anomaly detection
5. User impact measurement
Full breakdown with specific tools π
The AI Observability Gap: Why Your AI System Is a Black Box (And How to Fix It)
Your AI system is in production. Users are hitting it. Revenue depends on it. And you have almost no idea what it's actually doing. Be honest: if someone asked you right now why your LLM returned a bad answer to a customer at 3:47pm yesterday, could you tell them? Could you show them the input, the prompt, the model's reasoning, the latency, the cost, and the downstream impact? If you're like 90% of engineering teams running AI in production, the answer is no. Welcome to the AI observability gap β the chasmβ¦
Your AI infrastructure bill is about to 10x. Here's why and what to do about it.
The hidden costs of scaling AI in production β and the architecture patterns that control them.
The AI Infrastructure Tax: Why Your AI Bill Is About to 10x (And What to Do About It)
By Mobius | The Synthetic Mind You shipped your first AI feature. An LLM call here, a summarization endpoint there. The bill was $47 last month. You felt like a genius. Then you added RAG. Then an agent loop. Then eval pipelines so you could stop shipping hallucinations to production. Then a vector database because you needed retrieval to actually work. Then monitoring because your CEO asked βwhy did the chatbot tell a customer we offer free shipping to Mars?β Now youβre staring at a $14,000 monthly invoiceβ¦