The Luca Berton Blog
1937+ expert articles on AI, Kubernetes,Ansible, Platform Engineering, and cloud infrastructure — practical, hands-on guides and industry insights fromLuca Berton, AI & Cloud Advisor. Explore tutorials, deep dives, and the latest fromevery topic.
GA4 Key Events and Custom Dimensions as Code
Create GA4 key events and custom dimensions on every property from GitHub Actions with the Admin API, and avoid the three credential mix-ups I hit on the way.
Revolut Self-Hosts GLM on 56 B200s at $0.30 per Million Tokens
Revolut serves GLM-5.3 to 2,000 engineers on 56 NVIDIA B200s at $0.30 per million tokens, a tenth of its old API bill. How it tuned and governs the setup.
11 min read
OpenTelemetry, Turned Up: Auto-Instrumentation and Hot Sauce at Dash0
SRE NL at Dash0 in Amsterdam: Michele Mancioppi on every way to auto-instrument with OpenTelemetry, then an AMA with hot sauces on cost, semconv and Weaver.
7 min read
Arch9 2026: The Bottleneck Moved, and the Roads Under AI
Arch9 2026 by Levi9 in Amsterdam: review as the bottleneck, a platform under every AI agent, Scania's trucks, DPG Media's paved mesh and two banking stories.
12 min read
SREday Amsterdam 2026: Agents, GPUs and Five Conversations
SREday Amsterdam 2026 at ING Cedar: keynotes on watching AI agents, a GPU curled into a Kubernetes cluster, my main-stage talk and five Udienza conversations.
14 min read
Dutch Cloud Native Day 2026: Sovereignty, GPUs and Community
Dutch Cloud Native Day 2026 in Utrecht: sovereignty scored in eight dimensions, GPUs sliced between tenants, agents on stage, and six communities at one booth.
7 min read
Securing AI Agents in Four Layers: Lessons from AGNTCon
Identity, policy, observability and containment: how WSO2, Arcade.dev, Airia, Diagrid, AgentCat, Mezmo and Akamai secure AI agents, from AGNTCon San Jose.
7 min read
Techorama Netherlands 2026: Five Conversations in Utrecht
Techorama's medieval edition in Utrecht: Luise Freese on the proof-of-concept prison, Avalonia UI, school schedules at scale, AI pentesting, and 1,300 people.
6 min read
Apsara Conference Amsterdam 2026: Qwen, Routing and NVIDIA
Alibaba Cloud's first Apsara Conference in Europe: a new Amsterdam cloud region, one Qwen API that routes between models, and NVIDIA on agent runtimes.
14 min read
AGNTCon North America 2026: Every Takeaway
Every takeaway from AGNTCon + MCPCon North America 2026 in San Jose: the week's three, then each of my 25 conversations for Udienza, guest by guest.
28 min read
Diagrid Brings Durable Execution to goose, LangChainGo and Mastra
Diagrid's AGNTCon news: a Catalyst integration lets goose agents survive crashes and resume at the last committed step, plus LangChainGo and Mastra.
3 min read
AGNTCon North America 2026 Day 1: MCP Tasks and Skills
My day one at AGNTCon + MCPCon North America in San Jose: MCP Tasks in Claude Code, Skills over MCP, elicitation, and an AAIF of 305 members and 10 projects.
13 min read
AGNTCon North America 2026 Day 2: Small Models, Swap Drills
Day two of AGNTCon + MCPCon North America: Tim O'Reilly on MCP, small specialist models beat a frontier one, the swap drill, MCP at 16x, and 14 interviews.
10 min read
Building an AI-Native Startup with Band Desktop, Cognee, Flexprice and Snyk
Build an AI-native startup with agent collaboration, persistent memory, usage-based billing and automated security.
7 min read
TalosCon 2026 Day 1: Talos 1.14, Hypervisor, Talos Cupar
From Hotel Okura: Talos 1.14 with 44 features and no reboots, Talos Containers, the Hypervisor alpha, Talos Director, Talos for AI, and Yardi on stage.
14 min read
Nordic APIs 2026 Day 2: A Valid JWT Is Not Trust
Shared Signals and CAEP for revoking JWTs in flight, Arazzo workflows for agents, and my own talk: my notes from day 2 of Nordic APIs Summit in Stockholm.
11 min read
Prove GPUDirect Storage Works on NFS over RDMA
How to prove GDS is really moving data over NFS over RDMA instead of silently falling back, and why Run:ai pods still showed NFS as unsupported.
8 min read
Nordic APIs 2026 Day 1: Your Next Customer Is an Agent
UnConference sticky wall, Curity, Axway and Nordic APIs keynotes, agent identity, MCP and fuzzing: my notes from day 1 of Nordic APIs Summit in Stockholm.
16 min read
40 AI Web Design Tools, UI Libraries, and Inspiration Websites for 2026
Explore 40 AI web design tools, prompt libraries, UI components, animation resources and inspiration sites for building standout websites in 2026.
7 min read
Agent-Friendly Docs: llms.txt and Markdown with Hugo
How to make documentation readable by AI agents: the llms.txt format, Markdown versions of pages, and a working Hugo setup, inspired by Docsy's keynote.
5 min read
Agent Harness Fundamentals: Building Reliable AI Agents with Claude Code
Join Luca Berton and Youssef Hosni for a free Claude Code webinar on agent harnesses, hooks, subagents, automated tests, guardrails, and human review.
5 min read
Flatcar Container Linux: Immutable Kubernetes Nodes Guide
Why the host OS still matters in Kubernetes, and how Flatcar Container Linux delivers immutable, auto-updating nodes with Ignition, A/B updates and sysexts.
5 min read
Greg Kroah-Hartman on AI Bug Finders, the CRA and Linux at 35
My Udienza interview with Greg Kroah-Hartman in Prague: AI tools finding kernel bugs, kernel size per device, RISC-V, the CRA as an ingredients list, careers.
7 min read
Konstantin Ryabitsev on Email, b4 and the AI Patch Flood
My Udienza interview with Konstantin Ryabitsev in Prague: why Linux still runs on email, b4 and patch attestation, and the AI-generated patch flood.
7 min read