Arize AI
@arizeai
Followers
4K
Following
591
Media
496
Statuses
1K
Arize AI is an AI observability and LLM evaluation platform that helps teams deliver and maintain more successful AI in production.
Berkeley, CA
Joined January 2020
Today's the day!🎉 Arize Observe just kicked off, and it's bringing a whole set of new product announcements. From Agent-powered trace debugging to new Prompt Learning techniques, we've got it all! Announcements in the thread below 🧵 👇
1
6
18
Expertise around evals are rapidly becoming a must-have capability for product managers; our head of product @_amankhan can get you up to speed. Join us @vapi_ai HQ along with @producttank and @mixpanel. https://t.co/JVZZFr2N6C
0
0
3
Ready for day one of @mlopsworld in Austin! Stop by and say howdy, and check out our own Nick Luzio at 1:35pm in salon B for a rapid-fire demo & agent engineering tips.
0
1
5
You can now configure and run evals directly in the Arize UI at every level of your LLM application: span, trace, and session. 🔹 Span-level: Evaluate the smallest units of work such as retrieval quality, routing accuracy, or QA correctness. 🔹Trace-level: Understand the
1
0
2
We're hosting a workshop with @PagerDuty on how to stay prepared and responsive in high-stakes situations with AI agent deployments. RSVP:
pagerduty.com
Join PagerDuty and Arize as they unpack the real-world challenges organizations face in accelerating AI initiatives and share actionable strategies for success.
0
1
4
@OpenAIDevs just dropped Agent Builder, making it easier than ever to spin up and deploy agents. But once you’ve built an agent, how do you actually understand what it’s doing? Because these agents are powered by the @OpenAI Agents SDK, they can be traced with @ArizePhoenix or
1
5
11
Hope to see you at our #SFTechWeek happy hour on Tuesday with @Get_Writer
@basetenco @alloyautomation. Party + power panel -- and good food! RSVP:
partiful.com
As AI moves from pilot to production, two challenges stand above all others: how do we build reliable, scalable systems, and how do we instill trust at every layer of the stack? Join hosts WRITER,...
0
0
5
Santosh Vempala of @GeorgiaTech, will be joining our next Community Paper Reading to discuss his latest research paper with @OpenAI titled “Why Language Models Hallucinate.” RSVP: https://t.co/1cgaz173uj
0
0
2
Booking has AI agents deployed across its marketplace, which connects millions of travelers with memorable experiences every day. The latest in our "Rise of the Agent Engineer" series explores the how they built it:
0
0
3
Solid WORKSHOP happening tomorrow with @schavalii covering when to use reasoning, CoT, and explanations for LLM-as-a-Judge. RSVP: https://t.co/kErnTwcl26
0
5
9
Arize AI was just recognized in the IDC MarketScape: Worldwide GenAI Evaluation Technology Products 2025! We'd love to show you how Arize AX helps teams evaluate & ship agents faster: https://t.co/syUD04SLQa
0
1
2
LONDON: join @GoogleDeepMind and the Arize team for an evening of technical talks and demos focused on building and shipping reliable AI agents! RSVP required https://t.co/MwRHinKAwt
0
0
4
Two SOLID events with @aws for AI engineers next week (RSVP required; NYC one almost full): 🌉 SF (9/30): Agentic AI In Action feat @aws @mistralai @mongodb and others https://t.co/4CWVELyXS6 🏙️ NYC (10/2): Evaluating and Observing Agentic Workflows https://t.co/Jmsb2wXL7Z
0
0
6
With prompt management, you can: - Version prompts - Parametrize inputs - Replay prompts on historical data - A/B test in production - Compare models Think of it as Git for your prompts. A single place to manage, test, and evolve them. Example: @arizeai
1
1
2
Thrilled to reveal that @jason_lopatecki , CEO & Founder of @arizeai , will be joining our @awscloud Agentic AI Showcase! He’s tackling the big question: "How do you actually BUILD and SCALE AI agents for the enterprise?" 🔗 Grab your free ticket: https://t.co/kDtWqzWKAM
0
1
3
We are excited for what's possible with @dify_ai and Arize 🤝 Building AI agents is fast & intuitive with Dify, but keeping them accurate and reliable at scale can be a challenge. That’s where Arize comes in: trace every agent step, debug failures, run structured evaluations,
Dify 🤝 Arize We are excited to share how we bring observability into Dify apps with @arizeai. - Trace every step - Debug faster - Keep production agents reliable at scale With Arize Phoenix, developers can trace every agent call, debug failures, and run structured evaluations
0
3
9
Arjun Mukerji, PhD, of @atroposhealth will be presenting his paper on LLM summarization of real-world evidence studies at our next community paper reading! RSVP: https://t.co/oKMJgBwmh8
0
1
3
BERLIN: join @dat_attacked at @qdrant_engine Vector Space Day, where he'll be covering how to build self-improving evals for agentic RAG. RSVP: https://t.co/O6GizrSoPs
0
3
5
If you're debating whether to make the jump, our own Alec Swanson tackles the major differences between @cursor_ai and @Claude_Code and some power user techniques for the latter. https://t.co/Hpk5DdZneN
1
2
3
Writing & running run evals should feel smooth. We were able to spin this one up and run it in < 2 mins in the Arize UI! The new preview variables feature makes it simple to format LLM eval prompts correctly and avoid any frustrating errors. You can pull in any dataset
0
1
3
