Statsig Announces a New Azure AI Integration During Microsoft Ignite

BELLEVUE, Wash.--(BUSINESS WIRE)--In 2024, AI applications have grown up. While many companies were building AI applications in 2022 and 2023, few had achieved massive scale. Today, there are too many to count.

This shift has created new challenges for developers. The frameworks, tools, and implementation techniques that worked for test applications don’t work when deploying AI applications in production. Product builders need tools that help them configure, measure, and optimize their applications – but historically, these tools have been immature, fragmented, or non-existent.

Now, that’s changed. Statsig has collaborated with Microsoft to create the Statsig <> Azure AI Integration, giving anyone building with Azure AI a complete toolkit to configure AI models, implement them in code, track model performance, run powerful experiments to optimize outcomes, and see all the results in real time.

Over the past few years, many AI companies have relied on Statsig to optimize their products and release AI features – including OpenAI, Anthropic, Notion, Figma, and Brex. Statsig customer Dave Cummings, who is the Engineering Manager for ChatGPT at OpenAI, summarized it this way: “At OpenAI, we want to iterate as fast as possible. Statsig enables us to grow, scale, and learn efficiently. Integrating experimentation with product analytics and feature flagging has been crucial for quickly understanding and addressing our users' top priorities.”

Now, those features are available to any Azure AI customer via a simple SDK and a powerful native integration.

Statsig’s Azure AI SDK simplifies the implementation of features like completions and embeddings in server applications, by providing a layer of abstraction from direct Azure AI API calls and giving developers a simplified framework for implementing Azure AI models. This unlocks a very high level of flexibility for engineering teams.

The Statsig Azure AI SDK will also automatically log metrics on model and application performance to a Statsig project, giving companies a simple, out-of-the-box method for tracking model cost, latency, and performance. Down the road, this enables advanced experimentation, testing, and release strategies. Satya Nadella, Microsoft’s CEO, shared a summary of the partnership in the Ignite keynote: “We are working with Statsig to configure and run […] fast A/B tests using different models.”

This is part of Statsig’s broader push towards building advanced features for AI and ML use cases.

Statsig’s CEO, Vijaye Raji, summarized Statsig’s AI strategy in a brief statement: “Statsig is providing the foundation tools that AI companies need to optimize their applications. Experimentation powered innovation in machine learning, and now it’s powering the next generation of AI companies. We’re so excited to partner with Azure on a modern suite of tools for AI developers.”

Statsig has provided a full overview of the integration at https://statsig.com/blog/azure-ai-annoucement.