How I used Launch Templates to deploy AI workloads elastically across GPU providers and finally avoided vendor lock-in

We run a mixed GPU inference stack — H100s, H200s, RTX 5090s depending on availability and cost at any given time. For about a year, every time we wanted to shift workloads between providers we were effectively rebuilding deployment configs from scratch. Not because the workloads changed. Because the configs were hardcoded to one provider’s infrastructure. This is the actual GPU vendor lock-in problem and it took us embarrassingly long to name it correctly. What we thought the problem was We tho