Two weeks ago, I got to lead a hands-on workshop at the AnitaB.org Grace Hopper Celebration in Chicago: “Demystifying RAG: Build a Retrieval-Augmented Generation App from Scratch.” More than 200 women packed the room to learn, build, and ask some incredibly thoughtful questions.
I designed the session for people who wanted to start building with AI instead of only hearing about it. We began with a familiar problem: language models can sound confident while being completely wrong. Then we built the solution step by step—chunking source material, creating embeddings, storing them for vector search, retrieving the most relevant context, and using it to generate a grounded response.
It was so cool to see so many people come out, open their laptops, and build something together. The energy in the room made the whole experience feel even more special.
By the end, everyone had a working study-buddy RAG app and a clearer mental model of how LLMs can use their own data. The best part was watching people move beyond the notebook and start imagining what they could build next: search over personal notes, tools for their communities, podcast and video retrieval, and entirely new multimodal ideas.
This workshop reminded me how much I love making technical ideas feel approachable. Huge thank you to everyone who joined, shared their learning journey, and stayed curious—and to the friends who helped throughout the session and showed up for me in every way.
Want to build it yourself? The workshop repository has the notebook, setup instructions, slides, and resources. You can also see the workshop recaps on Instagram, Google Students, and LinkedIn.
Happy hacking 💜