Hi everyone :wave: We just launched something I t...
# general
m
Hi everyone ๐Ÿ‘‹ We just launched something I think many of you will appreciate โ€” especially if youโ€™ve ever tried
pulumi import
on a big GCP project ๐Ÿ˜… Itโ€™s called StackForge โ€” and it instantly converts a live GCP project into: โœ… A clean Pulumi YAML file (
Main.yaml
) โœ… A fully hydrated Pulumi state file (
stackName.json
) โœ… Optional GCS upload โ€” no deploy or apply needed ๐Ÿ›‘ No Terraformer ๐Ÿ›‘ No dependency chaos ๐Ÿ›‘ No manual reverse-IaC ๐ŸŽฅ Demo (2 min): http://x-itm.com/demo.mp4 ๐Ÿ” Trial access: https://x-itm.com/stackforge-trial/ (Request access key) Would love to hear what you think โ€” and open to early feedback or testing if anyone here wants to try it.
f
the idea is good, but the demo is not very compelling. 90% of time it shows nothing. would love to see the generated files and waiting part to be trimmed it seems you want it to be a commercial offering? too unpolished at the moment to be that, so I wouldn't even go through the hoops of generating api key and I don't know if I'd pay for the polished version either. imo, in 2025 with all genai tools, the value of low-level yaml/tf/IaC is not very high.
m
Hi Dmitry โ€” really appreciate the honest feedback ๐Ÿ™ Youโ€™re absolutely right that the current demo could be tighter โ€” especially the idle time during scanning. Weโ€™re actually re-editing it to jump straight into the generated
Main.yaml
and
stackName.json
so people can see the output immediately. As for the API key: totally fair. The key isn't tied to payment โ€” it just scopes access during early rollout. We want to make sure we onboard teams who are actively testing real-world GCP stacks so we can learn and improve fast. On the GenAI vs. YAML point โ€” I agree that higher-level abstractions are evolving quickly. But most of the infra teams weโ€™ve worked with (especially at consultancies or in regulated environments) still need clean, explicit IaC baselines before applying higher-level automation. StackForge isn't just generating YAML โ€” it's resolving live dependency graphs, stitching references across services, and producing state that's deployable on day one. So yes โ€” it's commercial in intent, but it's also born out of real-world pain trying to migrate complex GCP estates to Pulumi under deadlines. If we didnโ€™t build it, weโ€™d still be burning weeks manually importing resources ๐Ÿ˜… If you ever want to try it without jumping through forms, Iโ€™m happy to DM you a trial key directly. Either way โ€” really grateful for the input, and thanks for taking the time to engage ๐Ÿ‘Š