Guiding Agent Behavior Meetup - MLOps Community
Join us for the next MLOps.community meetup in San Francisco in partnership with Aporia where we'll be talking about how to guide agent behavior - focusing on enforcing character, guardrails, and staying factually accurate.
Me: "Help me write an intro for this meetup page without using the word delve"
LLM: "Join us for an enlightening session where we'll explore how to effectively manage hallucinations and avoid awkward interactions in customer-facing RAG applications. As much as prompt engineering can help shape responses, adding too many specific guidelines can overly complicate prompts, making it harder for large language models (LLMs) to adhere accurately to all instructions. If reliability is your priority, simply crafting intricate prompts won't suffice. In our discussion, we'll introduce alternative strategies that enhance reliability without overburdening the prompt. Whether you're a developer, a product manager, or simply interested in the future of AI interactions, this talk will equip you with essential insights to ensure your applications perform optimally in real-world scenarios. Don't miss out on this opportunity to learn how to maintain control and quality in AI communications."
Making agents behave in a certain way is hard. The above, for example, sounds almost too canned and unnecessarily verbose. So what are the strategies for guiding agent behavior that stay on track, provide a human-like interaction, stay on company brand, and do not get into 1$ contracts to sell cars. Come join us on May 17th at the Microsoft Reactor, where we'll hear from a breadth of experts on agent behavior from guiding NPCs, Code-writing agents, to adding strong guardrails for agents.
Agenda
5:30-6:15 Networking
6:15-6:20: Community Intro
6:20-6:40: Nyla Worker from Convai talks about guiding NPC agent behavior.
6:40-7:00: Neha Sharma from Databricks will talk about challenges enhancing the Databricks Assistant through RAG and prompt refinements, diverse experimentation like Side by Side UX, and robust offline evaluation metrics.
7:00-7:25: Alon Gubkin, CTO from Aporia, talks about Mitigating Hallucinations and Embarrassing Responses in RAG applications
More information coming soon.
NOTE: PLEASE RSVP BY WEDNESDAY MAY 15th 5:00 pm. Due to building security updates, we might not be able to allow walk-ins, or registrations on Friday.
About the MLOps Community:
MLOps Community fills the swiftly growing need to share real-world Machine Learning Operations best practices from engineers in the field. While MLOps shares a lot of ground with DevOps, the differences are as big as the similarities. We needed a community laser-focused on solving the unique challenges we deal with everyday building production AI/ML pipelines. We’re in this together. Come learn with us in a community open to everyone. Share knowledge. Ask questions. Get answers.
Join our Slack: MLOps Community
Thanks to the amazing folks at Microsoft Reactor for being a community partner and hosting our upcoming events!!
Note: To comply with the building Security, we've added a few questions that have been requested of us. Thank you for understanding.