Ben Lorica and Paco Nathan on LLM User Interfaces, AI Alignment, GenAI in the Enterprise, and a GenAI Scorecard.
Subscribe: Apple • Spotify • Overcast • Pocket Casts • AntennaPod • Podcast Addict • Amazon • RSS.
This is our monthly conversation on topics in AI and Technology with Paco Nathan, the founder of Derwen, a boutique consultancy focused on Data and AI.
Interview highlights – key sections from the video version:
- LLM Interface Innovation: Anthropic’s Claude 3.5 and the Power of “Artifacts”
- Code Generation and the Evolution of AI-Assisted Software Development Workflows
- Beyond Infinite Scrolling: The Need for Innovative LLM Interfaces
- Navigating the Challenges of Generative AI: Risks, Opportunities, and the Need for Taxonomies
- Understanding the Nuances of Generative AI Risks: Proper Use, Misuse, and Attacks
- AI Alignment Platforms: A Centralized Approach to Managing AI Risks at Scale
- Demystifying Graph RAG: Components, Challenges, and the Importance of Domain Expertise
- The Importance of Experimentation and Infrastructure for Graph RAG and LLMs
- The Environmental Footprint of AI: Power Consumption, Resource Use, and the Need for Mitigation
- Evolutionary Scale’s ESMFold: Accelerating Scientific Discovery with AI-Powered Protein Design
- AI as a Catalyst for Scientific Breakthroughs: From Protein Design to Weather Modeling
- LLM Routers, AI in Mathematics, and the FTC’s Watchful Eye on AI Services
- Recommendations
Relevant Links:
- What Are Artifacts in Claude?
- AI Risks:
- What is an AI Alignment Platform?
- Why Your Generative AI Projects Are Failing
- The Highs and Lows of Generative AI
- RAG, Graph AI and GraphRAG
-
- Enhancing Q&A with Domain-Specific Fine-Tuning and Iterative Reasoning: A Comparative Study
- GraphRAG: Design Patterns, Challenges, Recommendations
- Knowledge Graph-Enhanced RAG
- What Could Go Wrong When We Start Using LLMs to Organize Knowledge? 7 Pain Points of GraphRAG
- Boosting RAG Systems with Knowledge Graphs: Early Insights
- Lessons From The AI Revolution in Mathematics
- LLM Routers Unpacked
- FTC: Succor borne every minute
- Ben’s Recommendations (summer books):
- Impossible City: Paris in the Twenty-First Century 🇫🇷 (Simon Kuper)
- Other Rivers: A Chinese Education 🇨🇳 (Peter Hessler)
- Private Equity: A Memoir 💹 (Carrie Sun)
- Paco’s Recommendation: Simulating 500 million years of evolution with a language model
Related content:
- A video version of this conversation is available on our YouTube channel.
- Previous monthly roundups with Ben Lorica and Paco Nathan
- Philipp Moritz and Goku Mohandas → Navigating the Nuances of Retrieval Augmented Generation
- Semih Salihoglu → The Intersection of LLMs, Knowledge Graphs, and Query Generation
- Philip Rathle → Supercharging AI with Graphs
If you enjoyed this episode, please support our work by encouraging your friends and colleagues to subscribe to our newsletter: