Sonoma AI with Wine

This is an in-person event! Registration required in order to get in.

Topic: Sonoma AI (and wine) Meetups for remote SF Bay Area tech workers

What we’ll do:
Have some food and wine. Hear three exciting talks about developing with AI platforms, tools, unstructured data, and generative AI.

​3:30 - 4:30 - Welcome/Networking/Registration
4:35 - 5:00 - Christy Bergman, freelance AI Dev Advocate - data science with AI Tools (AI Studio, OpenAI, Perplexity)
5:05 - 5:30 - Paco Nathan, Principal DevRel Engineer, Senzing - Catching Bad Guys with AI apps
5:35 - 6:00 - Allison Ding, Developer Advocate Data Science, NVIDIA – Scaling Data Processing for Training Large Language Models
6:05 - 6:30 - Networking

​​​Tech Talk 1: Unlock Data Science with the power of AI
Speaker: Christy Bergman, freelance
Abstract: With a background in Data Science, I’ll guide you through essential data science tasks—like clustering, prediction, and querying—using Python and AI tools such as Google AI Studio, OpenAI, and Perplexity. Learn how to harness AI-powered techniques for quick data science prototyping.

​​​Tech Talk 2: Catching Bad Guys using open data and open models for AI apps
Speaker: Paco Nathan, Senzing
Abstract: We'll show use cases for investigative graphs​ and downstream AI applications which leverage them, such as GraphRAG. In efforts such as countering money laundering (AML), ultimate beneficial owner (UBO - catching oligarchs), human trafficking, etc., there are open datasets based on whistleblowers' leaks from banks, law firms, and others involved in the netherworld of Dark Money. If you've seen the Netflix film "The Laundromat" or read Oliver Bullough's "Moneyland" exposé, you've already seen how terrible this hidden landscape is and the impact it has on our daily lives. In this talk we'll explore use of state-of-the-art (SOTA) large models for constructing knowledge graphs from watchlists ("bad guys"), corporate disclosure documents, investigative journalism articles, and other related data sources. In particular, we'll cover use of high-end entity resolution​ and entity linking​ to build highly accountable workflows which mitigate AI "hallucinations" while following evidence-handling procedures needed to bring truly terrible people to court.

​​​Tech Talk 3: Scaling Data Processing for Training Large Language Models
Speaker: Allison Ding, NVIDIA
Abstract: Advancements in large language models (LLMs) have enabled developers to create a variety of applications, such as code generation, language conversion, and text summarization. The effectiveness of these models depends on the quality of the data used for training LLMs. Today, data exists in various formats, such as PDF files, tables, and documents. This data can be processed, filtered, and synthetically generated to address data scarcity and enhance model performance. Leveraging NVIDIA’s NeMo Curator, we can streamline the creation and augmentation of high-quality synthetic datasets, optimizing them for downstream tasks that train and/or fine-tune LLMs. In this talk, we'll explore the key challenges and solutions for processing massive datasets at scale to support these LLM applications.

Who Should attend:
Anyone interested in talking and learning about developing Generative AI Apps.

Where:
This is an in-person event. Registration using this form is required to get into the event. Registration in advance will close 2 days before the event. Capacity 50. Sponsored by LUMA OPTICS.

Are you interested in speaking or sponsoring a Sonoma AI with wine event? 🍷 Let us know!

Image: created using FLUX.1 schnell with the prompt renoir boating party with robot in the redwoods.