688: Six Reasons Why Building LLM Products Is Tricky

Om avsnittet

Prompt injection, prompt engineering, context windows, and more: In this week’s Five-Minute Friday, Jon explains why anyone looking to build their own product leveraging LLMs should stop to consider these and three more issues before jumping in. Phillip Carter first outlined these six issues in his article “All the Hard Stuff Nobody Talks About when Building Products with LLMs”.

Additional materials: www.superdatascience.com/688

Interested in sponsoring a SuperDataScience Podcast episode? Visit JonKrohn.com/podcast for sponsorship information.

Om avsnittet

Fler avsnitt

1038: AI’s Contentious Math Breakthroughs

1037: Recursive Self-Improvement: The Eureka Machine, with Dr. Richard Socher

1036: Is AI Holding Kids Back from Learning?

1035: Workslop, Glass Cliffs and Inviting Your Fears to Tea, with Carly Taylor

1034: In Case You Missed It in September 2026

1033: Workslop: The Hidden Cost of AI-Generated Work, with Prof. Jeff Hancock and Dr. Kate Niederhoffer

1032: Agents Need 10x More Data Than Humans, with Salesforce’s CDO Michael Andrew

1031: Tokenomics: Why Your Agentic AI Bill Is Exploding (and How to Fix It), with Tyler Cox and Ish Shah

1030: Garbage In, Gospel Out: Why Agents Need Better Data, with Salesforce's Gaurav Pathak

1029: How AI Brought a Podcast Back From the Dead, with Linear Digressions’ Katie Malone