Around the Prompt

Join Logan Kilpatrick and Nolan Fortman for a discussion with Shreya Rajpal that covers the inception and evolution of Guardrails, a tool designed to enhance reliability in AI applications. She emphasizes the importance of AI validation, the challenges of moving from proof of concept to production, and the organizational buy-in required for implementing such tools. The discussion also touches on the role of open source in AI development, the competitive advantages it provides, and the parallels between self-driving technology and AI systems. Shreya shares insights on real-world use cases, the introduction of Guardrail Server, and the future of AI regulation, highlighting the need for benchmarks and the importance of understanding the risks associated with generative AI. A few of our favorite Sound Bites: "Guardrails started as a form of building reliability." "AI development is very much like traditional software development." "The long tail is brutal in machine learning."

Join Logan and Nolan in a deep dive into the world of AI coding with Quinn Slack, the CEO of Sourcegraph. Sourcegraph was founded in 2013 with the goal of solving the problem of code search. The founders, who had experience working with massive code bases, wanted to create a code search tool similar to Google for code. They started by building code search as their first product and made mistakes along the way, but ultimately built a product that many developers love. Over time, Sourcegraph expanded to include intelligent automation and AI capabilities. They believe that the future of code search and AI in coding is a unified platform that combines search, chat, and AI capabilities. The conversation explores the potential impact of AI on coding and software development. Quinn Slack discusses how AI can empower non-technical individuals to code and create software solutions. He emphasizes the importance of making AI tools work manually before introducing automation. The conversation also touches on the challenges of building AI interfaces and the need for context integration from various tools. Quinn expresses his hope for the continued development of local models and competition in the AI space.Highlights: "We wanted to have some kind of code search like Google for code." "Search and AI chat really blur together. From the user's perspective, what they want is a box that they can type shit into and it solves their problems." "GitHub has the world's code. Why would we want to compete against GitHub? And I think now they're actually seeing a bunch of sort of co-pilot level competitors." "I think you're going to see that increase... we could have everyone coding or at least conjuring up code." "You got to make it work manually first before you introduce any kind of magic." "It is very likely that we'll end up with like many... one person billion dollar companies."

Andrew Mayne shares his non-traditional journey into AI, from being a magician and illusionist to becoming a science communicator at OpenAI. He discusses his early experiences with AI as a child, his interest in robotics and AI, and his fascination with chatbots. Andrew also talks about his experience using AI to deceive great white sharks and how it led him to explore the capabilities of text models like GPT-2 and GPT-3. He emphasizes the importance of prompt engineering and the need to carefully craft prompts to get desired outputs from the models. Andrew Mayne emphasizes the importance of having a clear idea of the desired output and breaking down complex tasks into manageable steps. He shares his experience in teaching magic tricks and how it helped him in prompt engineering. Mayne discusses the evolution of prompt engineering and the challenges and hype surrounding it. He also talks about his personal tech stack and the tools he uses for writing, coding, and research. Mayne expresses his excitement about the accessibility of AI models and the potential impact of AI in education. He also discusses his concerns about deep fakes and the need for trust and authentication in communication.

In this episode, Logan and Nolan dive deep into building AI product's with Marily Nika, a lead product manager at Google. We explore how AI on its own is not a product. AI product managers act as a bridge between AI and user needs. The role of AI product managers is to solve the right problems for users by leveraging AI capabilities. The demand for AI product managers is increasing, with companies like Anthropic and OpenAI actively hiring for these roles. AI product managers need to be comfortable with technology and have a good understanding of AI concepts and options. They also need to collaborate with scientists and engineers to make informed decisions about technical approaches. And much more!