
Stop Tuning Prompts. Build a Harness.
In a large codebase, the model is the smaller variable. The harness around it does the work: CLAUDE.md, hooks, skills, LSP, MCP, subagents.
Engin Diri
11 min readPage 2 of 8

In a large codebase, the model is the smaller variable. The harness around it does the work: CLAUDE.md, hooks, skills, LSP, MCP, subagents.
Engin Diri
11 min read
GPU clouds, MLOps platforms, and AI-powered infrastructure tools, compared. What each one is good at, where it falls short, and how to pick in 2026.
Alex Leventer
19 min read
Mention @pulumi-neo in GitHub or @Neo in your Slack workspace to bring Pulumi Neo into the threads where your team works on infrastructure.
Pulumi Neo Team
3 min read
Schedule any Neo task on a cadence. Drift checks, provider freshness, and encryption audits run themselves and deliver results as pull requests.
Pulumi Neo Team
3 min read
Ewan Dawson, CTO of Compostable AI, on building a software factory where AI agents write the code. Seven rules learned across nineteen client deployments.

2 authors
11 min read
Pulumi Neo is now available on the command line. Chat with Neo or run headlessly with the combined context of Pulumi Cloud and your local environment.
Pulumi Neo Team
3 min read
Connect Pulumi Neo to services like Datadog, Linear, and PagerDuty through MCP, and to AWS, Google Cloud, Azure, and Kubernetes through Pulumi ESC.
Pulumi Neo Team
4 min read
Ten more platform-engineering workflows you can hand to Pulumi Neo: deployments, tickets, incident triage, IAM, migration, and continuous ops.
Adam Gordon Bell
13 min read
Why infrastructure as code is the natural substrate for AI agents, and what we're shipping today to make agentic infrastructure even better.
Joe Duffy
15 min read
A year ago, building an AI agent meant frameworks, RAG, and glue code. The Claude Agent SDK, Codex SDK, and skills replaced most of that middle layer.
Engin Diri
8 min readPage 2 of 8