What we're researching, building, and shipping. Heart the things you want most — it helps us prioritize.
Ideas we're weighing
Stop juggling cluster switch before every deploy. Your app declares where it lives, and miren deploy sends it there.
Exploring an S3-compatible object storage addon, so apps get buckets without running MinIO or bringing an AWS account.
Tell Miren what healthy means for your app. Checks you define, so traffic waits for warm-up and wedged processes get restarted.
You should be able to read every line that touches your packets. We will be open-sourcing the Miren Anywhere POP and tunnel.
Queued to build
Upgrading Miren shouldn't drop your traffic. Ingress moves to its own process so requests keep flowing while the server restarts.
Share a cluster without sharing everything on it. Per-app permissions scope each user to their own apps.
One command backs up everything the cluster knows about itself, so a lost server doesn't mean a lost cluster.
Know who did what on your cluster: deploys, env var reads and writes, and when they happened.
Actively building
Today the dashboard knows almost nothing about your apps. We're fixing that: per-app health, current versions, and deploy history for every cluster.
Disks are becoming the default way to store data on Miren. Setup gets simpler, and backups flow to miren.cloud automatically.
Recently delivered
Inherit what an upstream image declares — entrypoint, port, workdir — so packaged apps deploy without a Dockerfile wrapper.
Making miren.md easier for AI agents to read. Biggest wins: machine-readable entry points (llms.txt/.md), fixing rendering artifacts, and adding metadata…