vitramir

4 ms·

1. 1.▲
   Ask HN: Which AI agent runtime would you bet on for production in 2026?

   2 points by vitramir 4mo ago|1 comments
2. 2.▲
   by vitramir 8mo ago

   terraform cloud, argocd, vercel and supabase (modern stack for micro apps), sentry (doesn't have per project permissions), sendgrid, etc... What does your stack look like beyond Kubernetes and AWS? It’s hard to imagine everything there
3. 3.▲
   by vitramir 8mo ago

   There’s also a related issue: many services use per-project API tokens. When agents need access to multiple projects, you have to pass several tokens at once. Which often leads to confusion and erratic behavior, including severe hallucinati
4. 4.▲
   by vitramir 8mo ago

   We’ve ended up solving this with an OAuth mock. The main advantages for us: - The application code is exactly the same in prod and preview environments. The only thing that changes is the OAuth provider configuration (endpoints, secrets, et
5. 5.▲
   by vitramir 8mo ago

   This sounds like an interesting approach. I’m not fully sure I understand one thing though. Are you suggesting to completely disable authentication inside the app and rely only on the encrypted channel? If yes, that’s not always an option —
6. 6.▲
   by vitramir 8mo ago

   > Wouldn't the client credentials app be a good fit for this? Or do you need user consent/scopes? If OAuth is already part of the product, switching flows only for preview environments isn’t really an option. It introduces a se