Moving off Heroku is rarely a technical problem. Heroku gives you a way of working: push, review app, add-on, rollback. AWS gives you primitives. Teams budget for moving the workloads, then lose six weeks rebuilding the workflow around them.
We'll spend the first half on why that happens and on the four usual escapes, each of which solves something and breaks something else. Then we run the migration live. An agent drives it, we narrate what's being provisioned underneath, and you see where it needs a human and where it doesn't.
Bring a stack you're stuck on.
What you'll learn
- Why the six weeks happen: the parts of Heroku that are workflow, not infrastructure, and why they don't come across in a lift-and-shift
- The four usual escapes: what each one buys you and what it costs you
- The migration, live: an agent running it end to end, with the AWS primitives it provisions explained as they appear
- Where a human is still required: the decisions you should not hand to an agent
Speakers
Guillaume Da Silva - Staff Cloud Architect at Qovery
Romaric Philogene - CEO & Founder at Qovery
Details
- Date: September 24, 2026 at 5pm CET / 11am EST / 8am PST
- Duration: 60 minutes including live Q&A
- Format: Talk plus live migration, with Q&A throughout
This session is for engineers and engineering leads still on Heroku who want their own AWS account without giving up the workflow they have today.