Latent Space: The AI Engineer Podcast
swyx + Alessio
The podcast by and for AI Engineers! In 2024, over 2 million readers and listeners came to Latent Space to hear about news, papers and interviews in Software 3.0.We cover Foundation Models changing every domain in Code Generation, Multimodality, AI Agents, GPU Infra and more, directly from the founders, builders, and thinkers involved in pushing the cutting edge. Striving to give you both the definitive take on the Current Thing down to the first introduction to the tech you'll be using in the next 3 months! We break news and exclusive interviews from OpenAI, Anthropic, Gemini, Meta (Soumith Chintala), Sierra (Bret Taylor), tiny (George Hotz), Databricks/MosaicML (Jon Frankle), Modular (Chris Lattner), Answer.ai (Jeremy Howard), et al.Full show notes always on https://latent.space
Episodes
Mentioned books
29 snips
Nov 1, 2024 • 41min
In the Arena: How LMSys changed LLM Benchmarking Forever
Anastasios Angelopoulos and Wei-Lin Chiang, both PhD students at UC Berkeley, lead the Chatbot Arena—a pioneering platform for AI evaluation. They discuss the evolution of crowdsourced benchmarking and the philosophical challenges of measuring AI intelligence. Emphasizing the limitations of static benchmarks, they advocate for user-driven assessments. The duo also tackles human biases in evaluations and the significance of community engagement, showcasing innovative strategies in AI red teaming and collaboration, all aimed at refining how language models are compared.
271 snips
Oct 25, 2024 • 1h 14min
How NotebookLM Was Made
Raiza Martin, Lead PM for NotebookLM at Google Labs, and Usama Bin Shafqat, AI engineer on the same project, share their insights on creating engaging AI-driven audio experiences. They discuss the unique approach of making digital voices sound conversational, incorporating natural interjections and pauses. The duo highlights the importance of user feedback in shaping their innovative product and the challenges of balancing advanced AI capabilities with accessibility for everyday users. Their work aims to redefine how we interact with information through AI.
40 snips
Oct 19, 2024 • 57min
Building the AI Engineer Nation — with Josephine Teo, Minister of Digital Development and Information, Singapore
In this engaging discussion, Josephine Teo, Singapore's Minister of Digital Development and Information, dives into the country's groundbreaking AI strategy. She emphasizes the balance between innovation and governance while exploring the societal impact of digital development. Teo sheds light on the importance of coding literacy and continuous education to prepare the workforce for tech transitions. The conversation also touches on Singapore's healthcare innovation landscape and the need for collaborations to nurture a vibrant AI ecosystem.
104 snips
Oct 18, 2024 • 1h 12min
Building the Silicon Brain - with Drew Houston of Dropbox
Drew Houston, CEO of Dropbox and a pioneer in cloud storage, shares his unique journey with AI, coding over 400 hours with LLMs. He discusses how Dropbox is evolving into an AI-first organization, highlighting innovations like Files GPT and universal search with Dropbox Dash. Drew emphasizes the importance of aligning technology with user needs while navigating data privacy and ethical practices. He explores the transformative potential of AI as a 'silicon brain' for knowledge work, urging a balance between human creativity and AI capabilities.
379 snips
Oct 11, 2024 • 1h 57min
Production AI Engineering starts with Evals — with Ankur Goyal of Braintrust
Ankur Goyal, founder of Impira and former head of Figma AI, discusses the evolving landscape of AI engineering. He highlights the challenges in LLM Ops, emphasizing the importance of evaluations in systematic AI development. Ankur reflects on his journey navigating career pressures as a first-generation Indian student and shares insights from his experiences in tech startups. He dives into advancements in data extraction, the integration of software engineering principles with AI, and the significance of community feedback for product improvements.
16 snips
Oct 3, 2024 • 2h 9min
Building AGI in Real Time (OpenAI Dev Day 2024)
Sam Altman, CEO of OpenAI, leads a team of experts including Kevin Weill and Michelle Pokrass, who unveil groundbreaking advancements in AI at Dev Day. They discuss the game-changing **Realtime API**, highlighting its capabilities for voice interaction and function calling. Alistair Pullen shares insights on developing AI models for software engineering. The conversation also dives into the challenges of AI fine-tuning and ethical considerations in AI communications, envisioning a future where seamless human-AI interaction transforms productivity.
111 snips
Sep 27, 2024 • 1h 30min
Language Agents: From Reasoning to Acting
Harrison Chase, founder of LangChain, LangSmith, and LangGraph, teams up with Shunyu Yao, an AI researcher known for his work on ReAct. They discuss the evolution of AI language agents, highlighting advancements in reasoning and acting. Shunyu shares insights on the ReAct framework and its impact on decision-making in AI. They delve into the challenges of benchmarking and interactive problem-solving in coding agents, while also exploring the future of AI model scaling and user experience in customer service applications.
256 snips
Sep 20, 2024 • 1h 9min
The Ultimate Guide to Prompting
Sander Schulhoff, author of The Prompt Report and creator of LearnPrompting.org, dives deep into the intricacies of prompt engineering and AI safety. He discusses the extensive research landscape with over 1,600 papers on prompting, clarifying popular methodologies like chain-of-thought and zero-shot prompting. Schulhoff also tackles ethical dilemmas surrounding AI-generated academic work and introduces innovative AI tools to streamline research methodologies. Further, he covers advancements in multimodal prompting, illustrating the complexities and potential of AI interactions.
234 snips
Sep 13, 2024 • 2h 4min
From API to AGI: Structured Outputs, OpenAI API platform and O1 Q&A — with Michelle Pokrass & OpenAI Devrel + Strawberry team
Michelle Pokrass leads the API Platform at OpenAI and has an impressive background in building scalable platforms at leading tech companies. In their discussion, they delve into the significance of structured outputs in AI, emphasizing its reliability for developers. The conversation covers everything from the latest advancements in OpenAI's capabilities, including the O1 model, to navigating database challenges and enhancing user experience through innovations in their APIs. They also touch on the complexities of cognitive biases in decision-making related to AI.
8 snips
Sep 3, 2024 • 1h 5min
Efficiency is Coming: 3000x Faster, Cheaper, Better AI Inference from Hardware Improvements, Quantization, and Synthetic Data Distillation
Nyla Worker, a Senior PM at Nvidia with a background in optimizing AI models at Google and eBay, shares insights on dramatic advancements in AI efficiency and inference. The discussion highlights a staggering reduction in costs and time for training models, with examples like the Cerebras platform achieving unheard-of speeds. They delve into optimizing large language models and the revolutionary potential of 3D conversational AI technology. Worker also touches on the future of digital personas and their applications in various sectors, including healthcare.
