Arize AI

@arizeai

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

San Francisco, CA

Arize AI
@arizeai
20h
Your agent can return a plausible answer and still take the wrong path. That's why these failures are hard to spot. Sept 24, we're hosting with @googlecloud and @MongoDB. Gemini does the thinking. MongoDB holds the memory. Arize shows the trace. Live. luma.com/agents-in-prod…
3
Arize AI
@arizeai
Sep 14
Going to @WeAreDevs in San Jose? 🎮 AI Trivia and Game Night, Wed Sept 23, 5-7pm at Guildhouse. 2 min walk from the convention center. Food, drinks, games, and the DevOps + SRE crowd. Hosted with @pagerduty and @Box RSVP 👉 luma.com/prtcc7pw?utm_s…
Arize AI
@arizeai
Sep 10
Prove it end-to-end: experiment on the full agent before you ship a change
1
Arize AI
@arizeai
Sep 9
Why should your agent send thousands of records through an LLM just to redact credit card numbers? Code mode lets the agent write a program that calls tools and transforms data inside a sandbox, so intermediate records can stay out of model context.
Code mode: Why your agent should code
From arize.com
6
Arize AI
@arizeai
Sep 9
One of our interns took a coding-agent workflow that cost about $100 per run on a frontier model and got it down to roughly $15-$20. That meant figuring out where expensive reasoning actually mattered.
How I cut coding agent costs with model and harness routing
From arize.com
4