Katherine Druckman talks to fellow Intel Open Source Evangelist Ezequiel Lanza about building and deploying AI applications using Retrieval Augmented Generation. We break down RAG concepts and processes involved in providing additional context to general AI models for specialized use cases.

Guest:

Katherine speaks with Demetris Cheatham, the Chief of Staff to the CEO of GitHub, about her unique perspective on the open source landscape. The discussion covers her experiences in various sectors and the impactful 'All In' project created to elevate developers from underrepresented backgrounds. They highlight the significance of community, the power of relationships, and the pivotal role of natural language and AI in making coding more accessible globally. The talk also addresses critical challenges like the digital divide, funding for diversity programs, and the importance of evolving diversity, equity, and inclusion efforts in tech.

00:00 Introduction 00:26 Connecting Through Open Source 02:02 Role and Responsibilities at GitHub 05:06 Diversity and Inclusion Initiatives 09:16 Challenges in Computer Science Education 12:51 Equity and Systemic Change 16:21 The Journey to a Billion Developers 24:07 Building Relationships in Open Source 31:37 Final Thoughts and Takeaways

Ezequiel Lanza and Katherine Druckman from Intel's Open Ecosystem team chat with Daniel Whitenack, founder and CEO of Prediction Guard. They discuss the importance and implementation of guardrails for securing generative AI platforms and cover the operational challenges and security considerations of running AI models, the concept of responsible AI, and practical advice for integrating guardrails into AI workflows. Additionally, the conversation touches on multi-model integrations, open source contributions, and the significance of vendor-neutral frameworks in achieving a secure and efficient AI ecosystem.

00:00 Introduction 01:28 What is Prediction Guard? 03:31 Understanding Guardrails in AI 06:49 Security Risks and Responsible AI 13:30 Open Source and Model Security 19:00 Open Platform for Enterprise AI 20:26 Contributing to Open Source Projects 27:12 Final Thoughts

Guest:

Daniel Whitenack (aka Data Dan) is a Ph.D. trained data scientist and founder of Prediction Guard. He has more than ten years of experience developing and deploying machine learning models at scale, and he has built data teams at two startups and an international NGO with 4000+ staff. Daniel co-hosts the Practical AI podcast, has spoken at conferences around the world (ODSC, Applied Machine Learning Days, O’Reilly AI, QCon AI, GopherCon, KubeCon, and more), and occasionally teaches data science/analytics at Purdue University.