I think there should be a jest extension that if a service's full test suite averages >10m over a day's worth of builds, the whole testing directory is deleted and you have to start again. Maybe bonus points if it just deletes the code as well. "Oh your service is too complex to test holistically in 10m? Try again..."

0 reply

0 recast

4 reactions

Find the flaw. How could this break? Goal: development platform for 10+ devs Why? Client needs it. We build it + personally love building such infra

1 reply

0 recast

6 reactions

I am once again humbled by my continued role as the perpetrator of my own terrible coding. I was discussing a coding issue with Claude AI and it asked, “what are the values of the constants in your code”? I…I'm not calling the correct constants in my code. 😩 On StackOverflow I'd get flamed, roasted, downvoted, and my question deleted. Man, I love (helpful) AI.

1 reply

0 recast

6 reactions

Good graphql benchmarking tool https://github.com/the-guild-org/gateways-benchmark/tree/main

0 reply

0 recast

2 reactions

"Kubernetes is for big companies only" - shut up. team of 5 can do it. They seem great judging from their API docs + Hiring page. Would work there. https://www.workatastartup.com/companies/recall-ai

5 replies

0 recast

7 reactions

I don't know why I need to add members to /backend - I already had very little time to manage the channel. So, I added a few mods (that were mods before they got rid of mods and now mods are back) and if you want to be added to the channel, just reply here or DC one of us.

7 replies

0 recast

6 reactions

There's finally a growing discussion about moving away from PaaS into simple deployable fullstack services (thanks to @dhh of course). I'm so here for it. We have been deploying our services to bare metal using https://kamal-deploy.org/ since day 1

1 reply

0 recast

4 reactions

If you want to send messages like this, /courtyard may be the place for you

0 reply

1 recast

6 reactions

I gotta say, as boring as it sounds, I thoroughly enjoy creating GraphQL schema and mapping it to my EdgeDB functions.

0 reply

0 recast

1 reaction

I always like the content Matt puts out about go. https://www.bytesizego.com/blog/golang-time

0 reply

3 recasts

5 reactions

Do you prefer ArgoCD or fluxCD and why? https://github.com/argoproj/argo-cd https://github.com/fluxcd/flux2

1 reply

1 recast

2 reactions

This channel has rotted :(

1 reply

0 recast

3 reactions

Do you use buildpacks? Any experience with buildpacks vs Docker? https://buildpacks.io/

3 replies

0 recast

2 reactions

Have folks here played with NATS? https://www.youtube.com/watch?v=hjXIUPZ7ArM

5 replies

0 recast

2 reactions

This week we found out during some load tests that our metrics library (cloudwatch) could not keep up with the requests, eventually OOMing the pods. This is an inherent issue with push-based metrics vs pull-based ones (e.g. Prom). With pull-based metrics you do zero outbound API calls, just have the scraper hit your metrics endpoint at a rate that you control. Prometheus ftw!

1 reply

0 recast

3 reactions