Lightning Lessons

RAG Without the Lag: Interactive Debugging For RAG Pipelines

Hosted by Quentin Romero Lauro and Shreya Shankar

Tue, Sep 9, 2025

9:00 PM UTC (30 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

Featured in Lenny’s List
AI Evals For Engineers & PMs
Hamel Husain and Shreya Shankar
View syllabus

What you'll learn

The Hidden Debugging Crisis in RAG Development

Over 70% of critical parameter changes require hours of re-processing the data, killing developer momentum.

Interdependency Traps That Breaks RAG Systems

Learn how chunk size changes force prompt rewrites, query rewriting impacts retrieval, and other cascading effects.

How Engineers Really Debug RAG Pipelines

Discover why engineers always debug retrieval first, plus real-world patterns from 12 engineers for effective testing.

Why this topic matters

RAG has become the backbone of enterprise AI, with 86%+ of companies using LLMs using RAG. Yet developers are stuck in slow feedback loops—unable to quickly test chunking strategies or validate component changes. Quentin and Shreya's latest paper on the human-centered side of RAG reveals expert workflows and introduces techniques for rapid RAG iteration not found in popular commercial tools.

You'll learn from

Quentin Romero Lauro

Retrieval & RAG Researcher

Quentin Romero is currently a Research Assistant at the University of Pittsburgh and will be joining character.ai as a Member of Technical Staff Intern in May 2025. His work focuses on the intersection of Human-Computer Interaction (HCI) and Artificial Intelligence, with projects like "RAGGY," a visual interactive tool for RAG pipelines, and "BizChat," a tool for small business owners to generate business plans using large-language models.

Quentin has a strong research background with publications in CSCW and an upcoming paper in VLDB. His work has been recognized with numerous awards, including 1st place in the Kuzneski Innovation Cup and "Most Technically Elegant Hack" at HackMIT.

Shreya Shankar

PhD Student at Berkeley EECS

Shreya Shankar is a final-year PhD student in EECS at UC Berkeley. Her research focuses on building reliable and efficient AI-powered data systems: specifically, developing query optimization algorithms that make AI-powered data analysis both accurate and affordable, as well as interfaces that help users specify their intent and trust the results. Her work, including DocETL (VLDB 2025) and DocWrangler (UIST 2025), has been deployed across multiple domains such as law, climate, and finance, and appears in top venues like VLDB, UIST, and SIGMOD. Outside of research, she co-teaches "AI Evals for Engineers," a popular online course for LLM evaluation and testing, which has enrolled 1,500+ participants from 500+ organizations to date.

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.