Hi IndieHackers,
Today I am coming up with a new idea that particularly concerns developers: what do you think about having an enhanced view of all your Amazon AWS products?
I find that with the current interface, it is pretty much impossible to easily know a complete list of all the things that you have (ec2, s3, ebs,... ) Across many different regions, categorize, link, group and document the purpose of your different assets.
As people are more and more moving their infrastructure to such cloud providers, I think that would be very beneficial to have some way to directly access a documented and easy way all your AWS products.
Would that also be the case for you?
If yes, I have a few questions:
Thank you very much for your time.
Feel free to let me know if you have any questions.
If you are interested, but don't want to participate, feel also free to drop me your email. I can write you once this will be live.
Best
Re: sharing AWS creds - I suggest you look at using cross account IAM roles to allow your users to grant your product access to their AWS accounts. This is a pretty standard approach for 3rd party cloud integrations.
Would I pay for a better cloud console? Eh, probably not. I've been working in ops / SRE roles for five years and know my way around the AWS and GCP consoles as well as the CLI interfaces. If I need information about my cloud resources, I know where to find it fast. Maybe I'm not the target demographic, but if that's the case consider that as novice users grow into their roles their need for an improved / simplified cloud console will decrease over time.
Other AWS console complexities: I hate switching accounts. I have like 8 or 9 AWS accounts I need to switch between across several organizations and for whatever reason their login page doesn't play very well with my password manager (LastPass). It's a pain in the ass, but as cloud pain goes it's a modest pain in the ass I guess.
You mentioned one thing in passing that caught my eye though. "backuping your configuration" is an intriguing prospect. Capturing infrastructure as code (terraform or cloudformation) is a tremendous amount of work. If you could inventory a set of cloud resources and then redeploy them quickly and easily as a disaster recovery tool that might be an interesting value prop.
Your users would still need automation to do last-mile provisioning, like if you recreate my RDS instance I still need to run my db migrations. If you recreate my EKS cluster I still need to ship my workloads to it (though you could interrogate and back those things too but that's going farther than your initial idea).
Thanks for the amazing feedback. I'll definitely include it in the reflection. Many good remarks in it!