Arize AI

@arizeai

The AI engineering platform for teams shipping reliable AI agents and LLM applications. Also home to @ArizePhoenix.

San Francisco, CA

arize.com

Replies Reposts Media

Arize AI
@arizeai
1h
Does your agent need retrieval on every turn? At @Upstart, Signal in Arize AX automatically surfaced unnecessary retrieval API calls that were adding cost at scale. Find recurring issues in your own traces: arize.com/blog/debug-pro…
Arize AI
@arizeai
Sep 18
For all you Codex fans out there, we've just added support for Codex as the harness for your managed agents. Configure your managed agents with Codex as the harness, select your models, skills, attach repos, and define your task. Then let your managed agent rip analyzing your
2
Arize AI
@arizeai
Sep 18
There's been a lot of chatter about Jev from @typesafeai this week, and we immediately got curious about what it means for how we build agents. So we got to work and dug in. Here's the latest on Jev from @aparnadhinak - this one's worth your time. 👇
Aparna Dhinakaran
@aparnadhinak
Sep 18
Article
Will TypeSafe’s Jev change how we build AI applications?
This week the AI community was in uproar about Jev from @typesafeai, not just a new model but a new kind of model: one that classifies, scores, and routes but cannot write a sentence. The reason for...
Arize AI
@arizeai
Sep 17
How can an agent pass its evals and still fail in prod? A response can look correct while the trajectory contains a bad tool call, a useless retry loop, or a success claim for an action that never happened. Signal is built to find those patterns across production traces.
7
Arize AI
@arizeai
Sep 17
From prototype to production: building reliable long-running agents with Temporal + Arize