Super Data Science: ML & AI Podcast with Jon Krohn 867: LLMs and Agents Are Overhyped, with Dr. Andriy Burkov
104 snips
Mar 4, 2025 
Dr. Andriy Burkov, a best-selling author and AI influencer, shares his insights on the future of AI, particularly questioning the hype around AI agents and large language models. He discusses innovative chatbot designs that avoid common pitfalls like hallucination. Burkov also reflects on the journey of language modeling, the evolution of natural language processing, and how Talent Neuron leverages data to transform talent management. He emphasizes the gap between human cognitive abilities and AI, stressing the skepticism around the effectiveness of AI in real-world applications.
AI Snips
Chapters
Books
Transcript
Episode notes
Early Web Scraping
- Andriy Burkov's interest in text analysis stemmed from his early internet experiences in 1998.
- He built web scrapers to automate information gathering, a precursor to his current work.
LLMs at Talent Neuron
- At Talent Neuron, Andriy Burkov uses language modeling to analyze 700M job postings daily, making 95M predictions.
- LLMs automate tasks like skill extraction, industry prediction, and salary extraction in multiple languages.
LLM Limitations
- LLMs excel at "in-distribution" problems similar to their training data but struggle with "out-of-distribution" problems.
- Andriy Burkov advises caution before deploying LLM-based systems in production due to unpredictable errors.
