Founder Real Talk AI's Trust Architects: The Patronus Approach to LLM-Based Applications and Enterprise Adoption with Anand Kannappan & Rebecca Qian
In a world where enterprises are moving faster than ever to adopt generative AI solutions, Patronus AI offers solutions to scalably and reliably catch mistakes and evaluate the performance of large language models (LLMs).
In this episode of Founder Real Talk, host Glenn Solomon and guest host Dan Cahana sit down with AI visionaries and Patronus AI co-founders Anand Kannappan and Rebecca Qian to explore their journey from CS students at the University of Chicago to AI leadership roles at Meta and the reasons they felt compelled to start Patronus after recognizing the need for robust evaluation tools following the release of ChatGPT in late 2022.
Anand and Rebecca illuminate the challenges enterprises face in adopting AI and the ways they’ve built Patronus to mitigate challenges like hallucinations and brand alignment. Against the backdrop of rapid AI adoption at large companies, the founders share their vision for creating the enterprise standard for unbiased, independent evaluations.
KEY TAKEAWAYS
- AI Visionaries: Anand and Rebecca's extensive experience in AI and machine learning, including their time at Meta and the University of Chicago, laid the foundation for Patronus.
- Day -1 for AI: The founders see AI adoption as still being in the very early stages, describing it as "Day -1." Large companies are moving faster than ever to adopt AI technologies but face challenges in navigating the rapidly evolving landscape.
- Solving Critical AI Issues: Patronus addresses the challenges of LLMs, such as hallucinations, unexpected behavior, and brand misalignments, making AI deployment safer and more reliable for enterprises.
- API-First Approach: Patronus's API-first solution allows for seamless integration into various parts of the tech stack, offering flexibility and scalability for real-time and offline use cases.
- Enterprise Adoption: The rapid adoption of AI by enterprises, driven by the need for reliability and safety, highlights the importance and urgency of Patronus's solutions.
- Future of AI: Patronus describes “Scalable oversight" and how humans will learn to supervise increasingly intelligent AI systems
RESOURCES
• Learn more about Notable Capital: www.notablecap.com
• Learn more about Patronus AI: https://www.patronus.ai/
CONNECT
• Connect with Glenn Solomon on LinkedIn: https://www.linkedin.com/in/glennsolomon/
• Follow Glenn Solomon on Twitter: https://x.com/glennsolomon
• Connect with Dan Cahana on LinkedIn: https://www.linkedin.com/in/dancahana/
• Follow Dan Cahana on Twitter: https://x.com/dan_cahana
• Connect with Anand Kannappan on LinkedIn: https://www.linkedin.com/in/anandnkannappan/
• Follow Anand Kannappan on Twitter: https://x.com/anandnk24
• Connect with Rebecca Qian on LinkedIn: https://www.linkedin.com/in/rebeccaqian/
• Follow Rebecca Qian on Twitter: https://x.com/rebeccatqian
Produced by Saschi Productions
Music by Gus Culpo
CHAPTERS
00:00 Intro and Introductions
5:00 What does Patronus do?
6:55 What’s been some of the surprises in AI?
10:00 The Patronus Product
12:15 How are customers using Patronus?
15:27 Why is an API-first approach important?
16:45 How did you know this was the right place to focus?
19:40 Large companies of the world are moving faster than ever
26:40 Into the Future: What will you look for to know that AI has been successful?
29:00 How do customers plan to deploy this type of technology?
32:00 AI Predictions
35:00 What motivates the Patronus founders?
36:40 What advice would you give to founders looking to start a company in the AI field?