Product
Tom Harding
New in Pulumi IaC: Support for skipping a resource
Pulumi now supports excluding specific resources from stack operations, giving you more control and efficiency in managing your infrastructure
Tom Harding
Page 5 of 8
Product
Pulumi Discovery Team
Go from Cloud to Code in Minutes with Visual Import
Introducing Pulumi Visual Import—discover, codify, and generate clean infrastructure code with AI assistance.
Pulumi Discovery Team
Customers
Shaun Verch
How We Used Pulumi to Safely Migrate Oso's Global Infrastructure
Learn how Oso used Pulumi to safely migrate its global infrastructure, ensuring zero downtime and maximum leverage.
Shaun Verch
Tutorials
Mikhail Shilkov
AI-Assisted Infrastructure as Code with Pulumi's Model Context Protocol Server
Learn how AI assistants like Cursor with Pulumi's MCP server accelerate IaC workflows and improve developer experience.
Mikhail Shilkov
Product
Will Jones
Improved refresh and destroy experience for Pulumi IaC
Pulumi now runs your code before refresh and destroy operations, enabling dynamic credential updates and complex workflows
Will Jones
Product
2 authors
Announcing the Next-Generation of Pulumi Components: Enabling Infrastructure Abstractions
Pulumi Components enable you to create, share, and consume reusable infrastructure building blocks across your organization and the broader community.
2 authors
General
Aaron Kao
Why Choose Pulumi Cloud Over DIY Backends?
Discover why to use Pulumi Cloud versus DIY IaC backends. Save time, reduce costs, and ensure compliance.
Aaron Kao
Engineering
2 authors
A Recipe for a Better AI-based Code Generator
Learnings from building a RAG for Pulumi code generator
2 authors
General
Adam Gordon Bell
Top 15 Python Tools for DevOps
From quick fixes to scalable enterprise solutions—explore 15 essential Python tools for automation, monitoring, and cloud deployment.
Adam Gordon Bell
Engineering3 authors
AI Engineering Lessons from Building Pulumi Copilot
Learn key engineering lessons from building Pulumi Copilot, including how to minimize LLM workload, validate outputs, and deal with hallucination.
3 authors
Page 5 of 8