Welcome to the 2025, tamuhack / Arize AI hackathon!

Howdy, Aggies! 👍 We're thrilled to have you join us for this exciting exploration of the cutting edge in artificial intelligence. If you've ever wondered about the tech behind game-changing products like ChatGPT, Cursor, and other incredible AI tools—or even secretly wished AI could do your homework (don’t worry, your secret’s safe with me : P)—this is your moment to dive in!

Generative AI is transforming the way we create, learn, and solve problems. It's the engine behind apps that write code, design art, and even hold meaningful conversations. Imagine harnessing this power to build something that could change lives—or just blow everyone away at your next project showcase. The possibilities are endless, and they’re at your fingertips.

This hackathon is more than just a competition. It’s a chance to experiment, collaborate, and spark ideas that could shape the future of AI. Who knows? Maybe you’ll build the next great product that defines this era of innovation.

So, gear up, dream big, and let's make some AI magic happen. The future is bright—and it's yours to build. Let’s do this! 🚀

Who Are We?

We’re Arize AI, a fast-growing startup based in Silicon Valley, and we live and breathe artificial intelligence. Ever wonder how AI forward companies like Uber/Doordash/Spotify can tell if their AI application is performing the way it’s supposed to? That’s the problem we solve.

At Arize, we’re pioneers in AI observability—creating software that monitors, evaluates, and troubleshoots AI models and applications. One of our products, the Arize AI Copilot, is like an AI detective whose sole purpose is to help other AI systems work better. Think of it as AI for AI—it doesn’t get cooler than that

I’m Dat, and I’m especially excited to be here because I’m one of you—I graduated from Texas A&M back in 2014 (not in computer science though). I’ve always wanted to give back to the Aggie community, and this hackathon is my way of doing that. While schools like Berkeley and Stanford in Silicon Valley have tons of exposure to the AI space, I felt it was important to bring that same energy and opportunity back to Texas. Y’all deserve to be right at the forefront of this technology revolution too.

Thank you so much for opening this quick start guide. I can’t wait to see what you build—Gig ’em!

What is Arize Phoenix?

Today you’ll be using Arize Phoenix

Phoenix is an open-source AI observability platform designed for experimentation, evaluation, and troubleshooting. It provides:

• Tracing - Trace your LLM application's runtime using OpenTelemetry-based instrumentation.
• Evaluation - Leverage LLMs to benchmark your application's performance using response and retrieval evals.
• Datasets - Create versioned datasets of examples for experimentation, evaluation, and fine-tuning.
• Experiments - Track and evaluate changes to prompts, LLMs, and retrieval.
• Prompt Playground - Experiment with different prompts to engineer the compare between different prompt templates

Phoenix is OpenTelemetry compatible, vendor and language agnostic with out-of-the-box support for popular frameworks (🦙LlamaIndex, 🦜⛓LangChain, Haystack, 🧩DSPy and more) and LLM providers (OpenAI, Bedrock, MistralAI, VertexAI, LiteLLM, and more).