Qovery

Infrastructure control plane for your coding agents. Agents deploy, you stay in control.

[Qovery @Qovery_]

You missed our “Heroku to AWS in One Command” live session yesterday - no problem. Here’s the summary and recording: https://t.co/GHefOWFExf

[Qovery @Qovery_]

4/ When do you pin one service back while the rest moves forward? A hotfix, a schema migration, a flaky dependency?

https://t.co/wz2V045ama

[Qovery @Qovery_]

1/ Roll one service back to a known-good commit while the rest of your environment keeps moving. Deploy by version lets you pick the exact version of each service and ship them all in one deployment. 🚀

[Qovery @Qovery_]

3/ Mutable "latest" tags can't be picked, so what runs is exactly what you selected. See pipeline takes you straight to the deployment.

[Qovery @Qovery_]

4/ It never merges its own PRs, and we haven't measured the time saved yet. Which alerts would you let an agent archive alone?

https://t.co/D7QTbP0Vp7

[Qovery @Qovery_]

1/ Every Sentry alert at Qovery gets a first look from an AI agent before an engineer picks it up. 🔎

[Qovery @Qovery_]

3/ One alert was a crash when sorting a table. The agent traced it to a missing value passed to localeCompare, opened a draft PR, and said it couldn't run the tests.

[Qovery @Qovery_]

4/ The safe pattern is propose, approve, apply. The agent gets read-only credentials and opens a pull request, a human or policy gate approves it, and a pipeline applies it. Which stage would you never hand to an agent?

https://t.co/9SUH3lZfGp

[Qovery @Qovery_]

1/ AI agents help with 4 stages of a cloud migration, and break in the 5 that decide whether it works. 🧩

[Qovery @Qovery_]

3/ They break on target sizing, data migration and cutover, network and identity, compliance evidence and the operating model. Those stay human decisions.

[Qovery @Qovery_]

4/ Also speaking: Thomas Humeau (Alan) and Charline Boulie (AWS). Seats are limited and registrations are reviewed: https://t.co/t6y5vJOWuS

[Qovery @Qovery_]

1/ In one week - Nov 12, Paris: "Agents in production: beyond the demo". Alan, AWS and Qovery host an evening of talks on running AI agents in production. 🧵

[Qovery @Qovery_]

3/ Romaric Philogène (CEO, Qovery) on "Who gives the agent access to production?": scope, check and contain, with a live demo of policy-based agent tokens.

[Qovery @Qovery_]

4/ When does a startup actually need its first DevOps hire? Give us the headcount or the trigger.

https://t.co/SO8iexDWGk

[Qovery @Qovery_]

1/ Hyperline moved off Heroku onto its own AWS setup, multiple environments included, within a few days and without hiring a DevOps engineer. ⚡

[Qovery @Qovery_]

3/ CTO Clément Garbay knew AWS, but nobody on the team was a DevOps specialist and a full-time hire was ruled out on cost. Qovery set up AWS and Kubernetes with best practices in place, and the developers manage infrastructure themselves.