Optimizing RAG: Leveraging Re-ranking and LLMs for Superior Results
Join us for an exciting evening of technical deep dives and practical insights into boosting search accuracy and relevance through advanced re-ranking techniques.
This is the second of a monthly series by crackedsf.com, a community focused on AI/ML engineering in the sf/bay area.
Featured Speakers:
Nicholas Khami – Founder & CEO at Treive
Nicholas is the driving force behind Treive, where he has successfully deployed re-rankers achieving low double-digit millisecond latency for search pipelines end-to-end. His expertise lies in optimizing the delicate balance between accuracy, latency, and customizability in re-ranking models. Nicholas will share his insights on navigating these trade-offs to enhance application performance.
Frank Liu – Head of Applied ML at Voyage AI
Voyage AI has deployed and commercialized some of the market's best re-rankers, and is the clear industry leader in code-specific and finance-specific domains. Frank will be speaking on improving the accuracy of RAG and semantic search.
Hubert Yuan – Technical Staff at Exa
As part of Exa's team of exceptional ML engineers, Hubert has contributed to building a pipeline for web-scale search. Currently, they're integrating re-rankers into their self-deployed production environment, pushing the boundaries of what's possible in large-scale search systems.
Michael Ryaboy – Developer Advocate at KDB.AI
With expertise in building LLM pipelines and teaching developers, Michael brings practical insights from his work at KDB.AI, where he focuses on optimizing search strategies. He’ll share an overview of re-ranking techniques and best practices from real-world experience.
Why Attend?
This is a unique opportunity to learn from industry professionals who are actively shaping the future of AI retrieval systems. Our speakers will offer deep dives into their experiences, sharing both the challenges and triumphs they've encountered in deploying re-ranking models in real-world applications. Whether you're an engineer, researcher, or AI enthusiast, their insights will provide valuable guidance for your own projects.
Don't miss out on the chance to:
Gain practical knowledge from experts who have successfully implemented re-ranking models.
Understand the trade-offs and considerations in optimizing re-rankers for performance and accuracy.
Network with professionals at the forefront of AI and machine learning innovation.
Agenda
5:30 PM - 6:00 PM: Registration and Networking
6:00 PM - 7:15 PM: Technical Presentations
7:15 PM - 9:00 PM: Q&A and Extended Networking
Each talk will include time for Q&A, allowing for in-depth technical discussions. However, please note that Q&A is limited to 5 minutes, though speakers will be available to answer questions afterward.
What to Expect
Technical deep dives with code examples and architecture discussions
Honest conversations about the challenges of scaling reranking
Opportunities to network with speakers and attendees facing similar technical hurdles
A collaborative atmosphere focused on knowledge sharing and problem-solving
Complimentary refreshments, including pizza, will be provided to fuel our discussions, provided by KDB.AI.
Space is limited because of our venue, so register now to secure your spot!
Call for speakers! We are seeking passionate individuals as speakers. If you have valuable insights to share with the AI/ML Infra community, please fill out this form: