Arize AI

Jason Lopatecki, Founder & CEO of Arize AI, delivered a keynote titled "Making GenAI Tractable: The Advantages of Speed and Scalability with Cerebras Wafer-Scale Clusters". As a second-time founder, Jason is a verified expert in AI, dig data architectures, and machine learning. His second venture, Arize AI is dedicated to helping engineers develop and improve models in one place. We're so excited to host such an experienced AI business owner! #GenAISummit2024 #GenAISummit

Webinar this week with Groq 👉 Building with LLMs and struggling to get the most out of your app? Just looking for more insight into the black box of modern AI development? Join us! John Gilhuly and Hatice Ozen will share a few hands-on examples that demonstrate how tracing and evaluation give you insight into every step of your LLM apps. Register: https://lnkd.in/gepEWJS7

Think you can build an AI agent worth bragging about? We’re kicking off Community Challenge Month with 30 days of prizes – including limited-edition Phoenix swag and a cash grand prize! 💰 Here’s how to enter... 1️⃣ Build an agent that’s truly dynamic – not just a predictable pipeline or chain. 2️⃣ Trace your agent with Arize Phoenix so it’s live in the Phoenix UI. 3️⃣ Code originality matters: submit code you’ve built, not just an example notebook from an existing repo! Use any tools you like, from LangGraph and Workflows to CrewAI and beyond. And if you missed our recent bootcamp, check out our course (in the comments) on building AI agents to get up to speed. Every eligible submission scores you Phoenix swag and a shot at the grand prize. 💸 Start building: https://lnkd.in/gEGQNR2C

🚀 Exciting update for developers building and monitoring AI applications! Microsoft just launched a new AI Apps Template Gallery—and Phoenix is officially part of it! This gallery is designed to provide developers with high-quality, easy-to-deploy templates that help you build great AI applications. With the new Phoenix integration, you can spin up a Phoenix instance hosted on Azure to bring observability into your workflow. Phoenix is open-source and free to self-host on Azure, making it an accessible option for any team. Here’s how it can support your AI projects: ✔️ Capture OpenTelemetry traces on AI applications, either manually or automatically through our dozens of integrations with libraries like LlamaIndex, Langchain, and Azure OpenAI ✔️ Run evaluations and experiments to measure hallucinations, context relevancy, function calling correctness, token usage, and more. ✔️ Curate datasets for streamlined fine-tuning and retraining of your underlying models Whether you’re focused on application monitoring, dataset curation, or performance evaluation, Phoenix in the AI Apps Template Gallery offers a structured, cost-effective starting point to boost your AI workflows. Check out the integration to get started: https://lnkd.in/gB3nqdZz

We're excited to be back in New York on Nov 19 🤠! Join experts from Google, LlamaIndex, Weaviate, AutoGen ( & Arize ) for an inside look at advanced techniques for building and optimizing agent systems. Register here: https://lnkd.in/dfi242BA

ICYMI: Check out the first session of our new series on building agents and assistants. Yesterday, Sally-Ann DeLucia and Aparna Dhinakaran shared their experience building Arize Copilot... They cover a lot of ground here including the Copilot architecture and router setup, including the AI Search skill. They walk through its initial development, share why they prioritized expanding this skill, and detail the iterative testing and evaluation process to enhance its performance continuously. https://lnkd.in/dybrx_cF

the best AI products winning in market right now are fun to use because they are 🔥 FAST 🔥 in this video i break down the design and engineering hacks that can turn your AI application into something magically responsive :)

If you are in NYC on November 19th, this is an event to not miss

Join us tomorrow at 10am PST, as Sally-Ann DeLucia and Aparna Dhinakaran kick off our new series on real-life agents in production! 🤖 The first session is all about Copilot, so we’ll be covering: 🏗️ Copilot’s architecture and router setup 🔎 Why expanding AI Search was a top priority 🔄 Our iterative testing process to boost performance In the rest of the series, we’ll cover real-life agents deployed in production, detailing their architectures, development systems, and lessons learned in real-world applications. Whether you’re a seasoned pro or just stepping into the agent world, we designed this series so there's something for everyone. Register: https://lnkd.in/gH3mcmPk

NYC folks: Really great opportunity coming up next month to learn from experts at Google, LlamaIndex, Priceline, Weaviate, AutoGen ( + Arize AI !) at our agents event. Some things that will be covered: - Advanced methodologies for evaluating AI agent performance in real-world scenarios - Proven methods for enhancing these systems to improve autonomy and robustness - New tools and frameworks transforming agent development and system evaluation - Best practices for iterating on AI behaviors with continuous feedback loops This is also a great opportunity to meet other people working on agents + assistants. https://lnkd.in/dy6qe8YU