Pulumi

Really cool to see Pulumi release provider for NVIDIA AI Cluster Runtime (AICR) for GPU software stack on #Kubernetes 👏 https://lnkd.in/g8NkpXJj

Estamos dando los primeros pasos para crear la comunidad de usuarios de Pulumi en Argentina! Si usas o usaste Pulumi (o estas interesado en Infracstructure as Code, cloud automation, agentic infra etc), me interesaria hablar con vos 🚀 La idea es organizar encuentros para compartir experiencias, desafíos y aprender unos de otros. Si te interesa sumarte, escribime por mensaje o dejá un comentario.

For about two years, the way you got something out of a coding agent was simple: write a good prompt, read what came back, write the next one. You held the tool the entire time. That part is ending. 🦄 Peter Steinberger says you shouldn't be prompting coding agents anymore. Boris Cherny, who heads Claude Code, says the same about his own job: "My job is to write loops." So you stop prompting the agent and start designing the loop that prompts it. The pieces already ship in Claude Code and Codex. The catch is the part that doesn't, and it's the reason two engineers can build the exact same loop and get opposite results. Engin Diri walks through what a loop is actually made of, and the part of it that no tool will ever do for you. https://lnkd.in/gv4qdzHJ

AICamp event reminder with two exciting announcements for FIFA raffle & Pulumi credits) ! Tomorrow, Pulumi returns to Silicon Valley with Laci Videmsky and Christopher Koning for another hands-on workshop on deploying AI agents on Amazon Web Services (AWS) using Amazon Bedrock AgentCore and Infrastructure as Code. If you're building AI applications and wondering how to handle authentication, secrets, IAM, observability, deployment, and scaling, this workshop is for you. Participants will be eligible for two bonuses - • Receive up to $300 in Pulumi credits to continue your learning and projects after the event • Enter a raffle to win a FIFA World Cup LEGO set sponsored by the Pulumi team June 9, 2026 | 5:30 PM PDT Plug and Play Tech Center, Sunnyvale RSVP: https://lnkd.in/gPCA2uwN Prerequisites: - Laptop with internet access - AWS account with Bedrock model access enabled (details provided ahead of the event) - Pulumi account (free tier works, sign up at [pulumi]) - GitHub account - Node.js 18+ or Python 3.11+ installed - Basic familiarity with the terminal Looking forward to another evening of learning, building, and connecting with the AI engineering community . #AIEngineering #AmazonBedrock #Pulumi #InfrastructureAsCode #AWS #AgenticAI #AIAgents #DevOps

Pulumi is in New York June 17 and 18, with three ways to get into deploying AI agents: a hallway chat at AWS Summit, a hands-on workshop, and the New York Pulumi User Group kickoff. Wednesday, June 17 Daytime: AWS Summit NYC at the Javits Center (429 11th Ave). Free, one day. Come find the Pulumi crew on the floor. Register at https://lnkd.in/dx65anB 5:30 PM: AI Camp Developer Workshop, "Deploying AI Agents with Pulumi and AWS" at GA, 10 East 21st Street (Flatiron). Engin Diri and Adam Gordon Bell take you through a working multi-agent system on Amazon Bedrock AgentCore. Auth, secrets, IAM, observability, all defined as code. The kind of thing you can actually run on Monday. Check-in 5:30, intro 6:00, hands-on 6:15, open Q&A 7:30. Hands-on, so seats are limited. Register on AI Camp: https://lnkd.in/gFWh8t_n Thursday, June 18 6:00 to 8:00 PM: the New York Pulumi User Group kicks off with Pinecone at 127 W 26th St (Chelsea). Theme: Scaling AI Infrastructure. Joerg Schad, VP Engineering at Pinecone, on how Pinecone builds infrastructure. Adam Gordon Bell on Momentum, the AI running coach he built with Pulumi. Hosted by Engin Diri. RSVP: https://lnkd.in/gEJ-Skzi Pick one. Pick all three. See you in New York.

We're a Silver sponsor at DevOpsCon Berlin, June 16-17. If you're attending, stop by our stand or join Wladi Mitzel's session, "Designed for agents, useful for humans: Five IaC practices that pay off twice." The 2025 DORA report found AI adoption now correlates with both higher delivery throughput and greater instability. The differentiator between teams that benefit and teams that don't isn't the AI. It's platform quality. AI just exposes the gap that was already there. Wladi walks through five practices that make platforms hold up for AI agents and humans alike, grounded in external research and Pulumi telemetry across thousands of organizations and nearly 10 million production deployments. Generalizable across Pulumi, Terraform, and OpenTofu. Looking forward to connecting with the platform engineering community in Berlin.

Many teams treat a git tag as the signal that a release is ready to ship. Pulumi Deployments now lets you act on that signal directly: configure tag-based triggers, push a version tag like v1.2.0, and Pulumi runs pulumi up automatically. Tag triggers work across all five version control integrations — GitHub, GitLab, Bitbucket, Azure DevOps, and Custom VCS — and use the same glob-based filtering model as path filters, so you can deploy on every v* release while excluding pre-releases like v1.2.0-rc1. The post also covers the PULUMI_CI_TAG_NAME variable for stamping the release version onto your resources, plus a setup note for existing GitLab integrations. https://lnkd.in/eKVyjk7T

Cloud APIs make AI agents easy to start. Then the tradeoffs surface: data leaves your network, offline work breaks, and every token is metered. Pablo Seibelt tested a different shape on a Mac: Gemma 4 through llama.cpp, Kubernetes for Open WebUI, and Tailscale for private access. The tricky part is not the model download. It is deciding which pieces belong on the host, which belong in the cluster, and how to keep the result reachable without turning it into another public endpoint. https://lnkd.in/eKXuX26u

I work for a Software vendor. And yes, we've added an Agentic AI part to our platform. But instead of being another chatbot that gives you suggestions and flatters how intelligent you are, this one actually understands your cloud environment and infrastructure. It can spin you up a new instance, help you manage routine infra tasks or go ahead and tag your infra however you like. It will do the boring repetitive work for you and so will help you get the work done, but it'll keep you in control and update you along the way. Now, your experienced human brain can get on with architecting, engineering and actually solving customer problems. Here's more from what Pulumi's Neo can do... Thanks Adam Gordon Bell for the write up! 💡 https://lnkd.in/eZm5QCGq #pulumi #infrastructureascode #agenticAIforInfrastructure

What happens when AI starts managing infrastructure, not just writing application code? In The Agentic Infrastructure Gap: In-Distribution Languages Make It a Coding Problem, Joe Duffy, Founder and CEO of Pulumi, will dive into why the future of infrastructure management may depend on modeling systems in the programming languages LLMs already understand best, like Python, TypeScript, and Go. This session will cover: • Why AI-written code will require AI-managed infrastructure • The limitations of infrastructure workflows built around manual processes and bespoke DSLs • How coding-language-native infrastructure creates better alignment for LLM reasoning • Building systems that map code changes directly to infrastructure outcomes • Lessons learned from working with leading AI companies and frontier labs preparing for agentic infrastructure operations As AI agents become increasingly capable operators, this talk offers a forward-looking perspective on how infrastructure platforms may evolve to support autonomous system management at scale. 📅 June 25, 2026 📍 Meydenbauer Center | Bellevue, WA Join us at @Scale: Systems & Reliability as we bring conversations around AI-native infrastructure and next-generation systems engineering to the Seattle/Bellevue area for the first time. Register now: https://bit.ly/4vfRfnB