One OpenAI-compatible API for hundreds of models on attested infrastructure, end to end encryption, with Confidential routes, ZDR routes, provider failover, BYOK, and no prompt or output logs.
@_joseph_perla_ End to end encryption means you can't see prompts or completions, which is the point, but it also takes away the logs I'd normally reach for when a response comes back wrong. What does debugging a bad request look like on your side?
@clement_avq you get to see all observability yourself and we’ve worked with customers g their ai agent debug recommendations
of course we have metadata about which provider we send to and content length so we’ve resolved issues at routing level not content level which is a different layer
Report
Really strong privacy angle here - the combination of confidential routes, ZDR, BYOK and provider failover makes this feel built for teams that need control and not just access. Congrats on the launch and all the best!!
Jammy Chat
Congrats on the launch, Joseph! 🚀
Atlantic Money
Congratulations on the launch!
@_joseph_perla_ End to end encryption means you can't see prompts or completions, which is the point, but it also takes away the logs I'd normally reach for when a response comes back wrong. What does debugging a bad request look like on your side?
TrustedRouter
@clement_avq you get to see all observability yourself and we’ve worked with customers g their ai agent debug recommendations
of course we have metadata about which provider we send to and content length so we’ve resolved issues at routing level not content level which is a different layer
Really strong privacy angle here - the combination of confidential routes, ZDR, BYOK and provider failover makes this feel built for teams that need control and not just access. Congrats on the launch and all the best!!