Since three years, Egor is bringing the power of AI to bear at Wise, across domains as varied as trading algorithms for Treasury, fraud detection, experiment analysis and causal inference, and recently the numerous applications unlocked by large language models. Open-source projects initiated and guided by Egor include wise-pizza, causaltune, and neural-lifetimes, with more on the way.
• Putting together a promising-looking prototype of multi-agent AI functionality on a local machine is easy, especially if you use one of the popular frameworks, such as LangChain or LllamaIndex. But how do you deploy and scale it in production? Do you need to use the corresponding framework’s paid offerings, such as LangServe and LangSmith?
• This talk will argue that AI agent orchestration is just a special case of distributed and event-driven workflows, which have been around for a long time. We will show how, for example, LlamaIndex’s Workflows semantics can be mostly replicated with Faust and Kafka; and briefly go over the many open source alternatives to LangSmith.
• Finally, we will describe the way Wise is going about productionizing and scaling its LLM workflows using Ray, a generic Python scalability framework.
Check out the incredible speaker line-up to see who will be joining Egor.
Download The Latest Agenda