DevOpsAgents – AI agents to deploy and manage your infra
AI agents with approval gates for infra changes, but execution-oriented AI is still unproven in production.

Waitlist-only IaC agent competing with Terraform and Pulumi.
Small teams without dedicated DevOps engineers
Terraform · Pulumi · AWS CDK
We've noticed a gap for small teams: getting a prototype working locally or on Vercel and Supabase is easy, but graduating to a full fledged deployment on a cloud provider is a nightmare.
If you don't have a dedicated DevOps engineer, you're usually stuck clicking around hoping that some combination of checkboxes will do the trick .
We're building Grafos V2 to act as your application's production agent.
• You provide your application, optionally tell it what you need if its not obvious (e.g. I need a special type of MLOps or custom model garden models) and choose your cloud provider.
• The agent understands the nitty-gritty details so you don't have to.
• It outputs verifiable infrastructure code, using your repository to store the source of truth.
We are opening a Closed Alpha to 100 users to test this out. If you're building anything and want your app productionized without the infrastructure headache, let us know in the comments or sign up in the link of the post
AI agents with approval gates for infra changes, but execution-oriented AI is still unproven in production.
Handles agent persistence, sandboxing, streaming—infra devs were rebuilding every time.
Another AI-powered deploy tool competing with Vercel, Railway, and Render.
Auto-synced infra-as-Markdown for Claude Code reduces agent context discovery time.
Hackathon-stage; needs real validation that repo analysis beats manual estimation.
LLM agent chains cloud CLIs and git logs, but needs runbook guardrails to avoid catastrophic mistakes.