Super Data Science: ML & AI Podcast with Jon Krohn 853: Generative AI for Business, with Kirill Eremenko and Hadelin de Ponteves
140 snips
Jan 14, 2025 
Join Kirill Eremenko, CEO of Super Data Science, and Hadelin de Ponteves, a leading AI educator, as they delve into the transformative power of foundation models for businesses. They discuss the eight-step lifecycle for implementing these models, offer criteria for selecting the best fit, and explore clever customization techniques. The duo also introduces AWS generative AI tools, making it easier for companies to leverage AI without breaking the bank. Their insights are a treasure trove for anyone looking to navigate the generative AI landscape!
AI Snips
Chapters
Transcript
Episode notes
Foundation Model Lifecycle
- Understand the eight-step lifecycle of foundation models: data prep, model selection, pre-training, fine-tuning, evaluation, deployment, monitoring, and iteration/maintenance.
- Large companies handle the first three steps; businesses typically focus on fine-tuning and subsequent stages.
Fine-tuning Example
- Hadelin and his students fine-tuned a LLaMA model using medical data from Hugging Face.
- They created a chatbot proficient in advanced medical terminology, demonstrating the power of fine-tuning.
Foundation Model Selection Criteria
- Consider twelve key criteria when choosing a foundation model: cost, modality, customization, inference, latency, architecture, benchmarks, language, size, scalability, compliance, and environmental impact.
- Prioritize based on your business needs and application requirements.