Episodes
Four years of tape. Every episode is checked claim by claim and playable.
64 episodes in 2024
🎙LATENT SPACE
51m Best of 2024 in Agents (from #1 on SWE-Bench Full, Prof. Graham Neubig of OpenHands/AllHands)
Professor Graham Neubig shares technical insights and architectural lessons from building OpenHands, an open-source autonomous coding agent framework. Through live demonstrations and researc…
25 statements 6 featured 3 assessments
🎙LATENT SPACE
28m Best of 2024: Synthetic Data / Smol Models, Loubna Ben Allal, HuggingFace [LS Live! @ NeurIPS 2024]
At NeurIPS 2024, Loubna Ben Allal of Hugging Face presents a comprehensive overview of synthetic data generation pipelines and the rapid advancement of small, on-device language models. She …
21 statements 7 featured 10 assessments
🎙LATENT SPACE
42m 2024 in Post-Transformer Architectures: State Space Models, RWKV [Latent Space LIVE! @ NeurIPS 2024]
At NeurIPS 2024, Dan and Eugene examine the emergence of post-transformer architectures, detailing how State Space Models (SSMs) and RWKV overcome quadratic attention bottlenecks through sub…
15 statements 4 featured 2 assessments
🎙LATENT SPACE
37m Best of 2024: Open Models [LS LIVE! at NeurIPS 2024]
Presented at NeurIPS 2024, this session reviews the major milestones, technical definitions, and systemic challenges facing open language models in 2024. The presentation features an in-dept…
16 statements 6 featured 5 assessments
🎙LATENT SPACE
55m Best of 2024 in Vision [LS Live @ NeurIPS]
Presented live at NeurIPS, this technical retrospective highlights the defining computer vision breakthroughs of 2024, analyzing generative video architectures, the ascendance of real-time D…
23 statements 5 featured 12 assessments
🎙LATENT SPACE
26m The State of AI Startups in 2024 [LS Live @ NeurIPS]
At Latent Space Live during NeurIPS, Sarah Guo and Pranav Reddy of Conviction present a comprehensive retrospective on the 2024 AI landscape, examining foundational model competition, test-t…
12 statements 4 featured 4 assessments
🎙LATENT SPACE
1h 6m Windsurf: The Enterprise AI IDE
In this podcast episode, Codeium co-founder Varun Mohan and VP of Product Anshul Ramachandran detail the creation of their native AI IDE Windsurf, discussing its underlying agentic architect…
30 statements 8 featured 3 assessments 2 contradicted
🎙LATENT SPACE
43m [Paper Club] Weight Streaming on Wafer-Scale Clusters (w/ Sarah Chieng of Cerebras)
In this Paper Club session, Sarah Chieng of Cerebras breaks down the weight streaming architecture for wafer-scale clusters, explaining how decoupling parameter storage from compute units ov…
10 statements 4 featured 9 assessments 1 contradicted
🎙LATENT SPACE
1h 36m 0 to over $8M ARR in 2 months as a Claude Wrapper (Bolt.new, Qodo)
In this episode of the Latent Space podcast, hosts Alessio Fanelli and Swix interview Qodo CEO Itamar Friedman and StackBlitz CEO Eric Simons to dissect the rise of AI coding agents, compari…
29 statements 9 featured 5 assessments 2 contradicted
🎙LATENT SPACE
41m [Paper Club] Embeddings in 2024: OpenAI, Nomic Embed, Jina Embed, cde-small-v1 - with swyx
In this Paper Club session, swyx leads an in-depth analysis of the 2024 embedding landscape, examining architectural innovations like Matryoshka learning, task-specific LoRA adapters, and co…
0 statements
🎙LATENT SPACE
55m [Paper Club] DocETL: Agentic Query Rewriting + Eval for Complex Document Processing w Shreya Shankar
In this Paper Club session, host Eugene Yan and author Shreya Shankar break down DocETL, an agentic framework that formalizes complex document processing through database-style operators, qu…
17 statements 5 featured 4 assessments 1 contradicted
🎙LATENT SPACE
1h 11m The new Claude 3.5 Sonnet, Computer Use, and Building SOTA Agents — with Erik Schluntz, Anthropic
Anthropic's Erik Schluntz discusses the engineering principles behind Claude 3.5 Sonnet's state-of-the-art SWE-bench performance, mistake-proof tool design, the launch of Computer Use, and p…
24 statements 7 featured 2 assessments
🎙LATENT SPACE
53m [Paper Club] BERT: Bidirectional Encoder Representations from Transformers
This paper club session presents an in-depth architectural and practical review of Google's landmark 2019 paper on BERT (Bidirectional Encoder Representations from Transformers). The partici…
0 statements
🎙LATENT SPACE
55m Why Compound AI + Open Source will beat Closed AI — with Lin Qiao, CEO of Fireworks AI
In this episode, Fireworks AI CEO Lin Qiao sits down with Alessio Fanelli and Swyx to discuss the technical architecture of distributed inference, the shift toward Compound AI and declarativ…
13 statements 6 featured 4 assessments
🎙LATENT SPACE
1h 8m Agents @ Work: Lindy.ai (with live demo!)
On the Latent Space Podcast, Lindy.ai founder and CEO Florent Crivello explores the evolution of AI agents from fragile prompts to deterministic no-code workflows, delivers comprehensive liv…
42 statements 12 featured 1 assessments
🎙LATENT SPACE
58m Agents @ Work: Dust.tt — with Stanislas Polu
Former OpenAI reasoning researcher and Dust co-founder Stanislas Polu shares inside perspectives on OpenAI's scaling culture and details the engineering and product architectures needed to d…
25 statements 6 featured 3 assessments
🎙LATENT SPACE
52m [Paper Club] Intro to Diffusion Models and OpenAI sCM: Simple, Stable, Scalable Consistency Models
In this Paper Club presentation, speaker RJ delivers a comprehensive breakdown of OpenAI's Simple, Stable, Scalable Consistency Models (sCM), explaining how continuous-time probability flow …
12 statements 4 featured 7 assessments 1 contradicted
🎙LATENT SPACE
41m In the Arena: How LMSys changed LLM Benchmarking Forever
In this episode of the Latent Space podcast, LMSYS researchers Wei-Lin Chiang and Anastasios Angelopoulos discuss the origins, statistical methodology, and expansion of Chatbot Arena into th…
17 statements 7 featured 5 assessments
🎙LATENT SPACE
39m [Paper Club] Upcycling Large Language Models into Mixture of Experts
Ethan from NVIDIA explains the principles of Mixture of Experts (MoE) architectures, details Megatron Core systems optimizations, and presents a methodology for upcycling pre-trained dense l…
8 statements 4 featured 3 assessments
🎙LATENT SPACE
1h 13m How NotebookLM Was Made
In this episode of Latent Space, Google Labs product lead Raiza Martin and AI engineer Usama Shafqat join hosts Swix and Alessio Fanelli to discuss the inception, architectural design, and v…
26 statements 8 featured 6 assessments
🎙LATENT SPACE
56m Singapore: the AI Engineer Nation — with Minister Josephine Teo
In this episode of the Latent Space Podcast, Singapore's Minister for Digital Development and Information, Josephine Teo, details the refreshed National AI Strategy 2.0, exploring workforce …
21 statements 6 featured 8 assessments
🎙LATENT SPACE
1h 1m [Paper Club] SWE-Bench [OpenAI Verified/Multimodal] + MLE-Bench with Jesse Hu
In this Paper Club presentation, Jesse Hu and host Eugene explore the architecture, evolution, and practical realities of SWE-bench, SWE-bench Verified, SWE-bench Multimodal, and MLE-bench. …
21 statements 9 featured 10 assessments
🎙LATENT SPACE
1h 11m Building the Silicon Brain - Drew Houston of Dropbox
Dropbox Co-founder and CEO Drew Houston joins the Latent Space podcast to discuss his hands-on AI engineering practices, the technical architecture behind production RAG systems, and Dropbox…
31 statements 10 featured 1 assessments
🎙LATENT SPACE
1h 12m [Paper Club] Molmo + Pixmo + Whisper 3 Turbo - with Vibhu Sapra, Nathan Lambert, Amgadoz
In this Paper Club session, speakers Vibhu Sapra, Nathan Lambert, and Amgadoz analyze the architecture and data curation pipeline behind AI2's open vision-language model family Molmo and Pix…
13 statements 6 featured 8 assessments
🎙LATENT SPACE
1h 56m Production AI Engineering starts with Evals
In this in-depth interview, Braintrust founder and CEO Ankur Goyal discusses the transition of AI development from academic data science to developer-centric software engineering, drawing on…
42 statements 11 featured 3 assessments
🎙LATENT SPACE
41m [Paper Club] Berkeley Function Calling Paper Club! — Sam Julien, Writer
Sam Julien presents the architectural evolution of the Berkeley Function Calling Leaderboard across three versions, leading into an interactive technical discussion on multi-turn evaluation,…
4 statements 1 featured 2 assessments 1 contradicted
🎙LATENT SPACE
2h 9m Building AGI in Real Time (OpenAI Dev Day 2024)
This comprehensive dispatch from OpenAI Dev Day 2024 explores OpenAI's major technical breakthroughs through in-depth interviews and live demonstrations. The coverage details the WebSocket-p…
23 statements 8 featured 7 assessments 1 contradicted
🎙LATENT SPACE
1h 0m [Paper Club] Who Validates the Validators? Aligning LLM-Judges with Humans (w/ Eugene Yan)
Eugene Yan leads a Paper Club discussion analyzing the paper 'Who Validates the Validators? Aligning LLM-Judges with Humans' by Shreya Shankar et al., exploring methodologies for constructin…
8 statements 2 featured 1 assessments
🎙LATENT SPACE
1h 26m Language Agents: From Reasoning to Acting — with Shunyu Yao of OpenAI, Harrison Chase of LangGraph
LangChain creator Harrison Chase and researcher Shunyu Yao join the Latent Space podcast to discuss the evolution, cognitive architectures, evaluation benchmarks, and tooling design powering…
30 statements 7 featured 3 assessments
🎙LATENT SPACE
23m llm.c's Origin and the Future of LLM Compilers - Andrej Karpathy at CUDA MODE
In this CUDA MODE presentation, Andrej Karpathy chronicles the development of llm.c—a minimalist, dependency-free framework for training transformer models in pure C and CUDA—and explores ho…
14 statements 4 featured 5 assessments
🎙LATENT SPACE
1h 7m The Ultimate Guide to Prompting - with Sander Schulhoff from LearnPrompting.org
Sander Schulhoff, creator of LearnPrompting.org and lead author of The Prompt Report, joins the Latent Space podcast to unpack the empirical science of prompt engineering, adversarial AI saf…
18 statements 6 featured 6 assessments
🎙LATENT SPACE
53m [Paper Club] Writing in the Margins: Chunked Prefill KV Caching for Long Context Retrieval
Machine learning researcher Umar Jamil presents 'Writing in the Margins,' an inference-only technique that leverages chunked KV cache prefilling in transformers to generate intermediate anno…
11 statements 4 featured 6 assessments
🎙LATENT SPACE
47m [Paper Club] 🍓 On Reasoning: Q-STaR and Friends!
This Paper Club presentation examines the progression of self-taught reasoning in language models, detailing the bootstrapping mechanics of STaR, the continuous token-level thoughts of Quiet…
1 statements 1 assessments
🎙LATENT SPACE
1h 12m Building AGI with OpenAI's Structured Outputs API
OpenAI Tech Lead Michelle Pokrass sits down with Alessio Fanelli and Shawn Wang to provide an in-depth technical look at the engineering, constrained decoding mechanics, and developer best p…
32 statements 9 featured 7 assessments
🎙LATENT SPACE
1h 7m Personal benchmarks vs HumanEval - with Nicholas Carlini of DeepMind
DeepMind research scientist Nicholas Carlini joins the Latent Space podcast to discuss pragmatic LLM workflows for developers, the necessity of personalized AI benchmarks over public leaderb…
31 statements 9 featured 8 assessments
🎙LATENT SPACE
1h 1m Is finetuning GPT4o worth it?
In this episode of Latent Space, Cosine CEO Ali Pullen joins Alessio Fanelli and Swix to break down the technical architecture, synthetic training pipelines, and large-scale OpenAI fine-tuni…
33 statements 9 featured 6 assessments 1 contradicted
🎙LATENT SPACE
1h 10m Answer.ai & AI Magic with Jeremy Howard
In this Latent Space podcast episode, Jeremy Howard discusses Answer.ai's Public Benefit Corporation mission, breaks down systems engineering breakthroughs for local fine-tuning, challenges …
25 statements 7 featured 2 assessments
🎙LATENT SPACE
1h 0m Segment Anything 2: Memory + Vision = Object Permanence — with Nikhila Ravi and Joseph Nelson
In this episode of the Latent Space Podcast, Meta FAIR lead author Nikhila Ravi and Roboflow's Joseph Nelson explore the release of Segment Anything 2 (SAM 2), detailing its architectural in…
19 statements 6 featured 10 assessments 1 contradicted
🎙LATENT SPACE
1h 23m The Winds of AI Winter (Q2 Four Wars of the AI Stack Recap)
In this quarterly recap recorded in Singapore, Latent Space Podcast co-hosts Alessio Fanelli and Swix analyze the 'Four Wars of the AI Stack,' assessing frontier model competition, hardware …
6 statements 1 featured 1 assessments
🎙LATENT SPACE
1h 23m [LLM Paper Club] Llama 3.1 Paper: The Llama Family of Models
In this Latent Space LLM Paper Club session, hosts and AI practitioners dissect Meta's Llama 3.1 research paper, analyzing architectural design, massive synthetic data pipelines, compute inf…
7 statements 1 featured 3 assessments 2 contradicted
🎙LATENT SPACE
1h 4m Training Llama 2, 3 & 4: The Path to Open Source AGI — with Thomas Scialom of Meta AI
In this episode of the Latent Space podcast, Meta AI technical lead Thomas Scialom discusses the engineering breakthroughs, scaling principles, and post-training innovations behind the Llama…
26 statements 6 featured 7 assessments
🎙LATENT SPACE
2h 18m The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
In this in-depth interview, former Google Brain researcher and Reka co-founder Yi Tay provides insider perspectives on transformer architectures, scaling laws, startup compute hurdles, and t…
34 statements 9 featured 3 assessments 1 contradicted
🎙LATENT SPACE
1h 32m State of the Art: Training 70B LLMs on 10,000 H100 clusters
In this episode of the Latent Space Podcast, Jonathan Frankle (Databricks) and Josh Albrecht (Imbue) discuss the low-level systems engineering, bare-metal hardware realities, and empirical o…
43 statements 12 featured 8 assessments
🎙LATENT SPACE
1h 8m How To Hire AI Engineers (ft. James Brady and Adam Wiggins of Elicit)
Elicit engineering leaders James Brady and Adam Wiggins break down the core competencies, architectural principles, and hiring strategies required to build effective AI engineering teams. Th…
15 statements 4 featured
🎙LATENT SPACE
1h 5m How AI is Eating Finance - with Mike Conover of Brightwave
In this episode of the Latent Space podcast, Brightwave founder Mike Conover discusses the technical and strategic realities of building vertical AI for financial services. He explains how m…
31 statements 8 featured 2 assessments
🎙LATENT SPACE
1h 12m How to train a Million Context LLM — with Mark Huang of Gradient.ai
Mark Huang, co-founder of Gradient.ai, joins the Latent Space podcast to break down how his team extended Llama-3 to a one-million token context window, discussing RoPE theta scaling, Ring A…
23 statements 6 featured 3 assessments
🎙LATENT SPACE
55m LLM Asia Paper Club Survey Round
The LLM Asia Paper Club survey session convenes researchers to present and evaluate recent developments in large language model reasoning, uncertainty estimation, mechanistic interpretabilit…
0 statements
🎙LATENT SPACE
1h 56m This World Does Not Exist — Joscha Bach, Karan Malhotra, Rob Haisfield (WorldSim, WebSim, Liquid AI)
This special three-part Latent Space episode explores the simulative frontier of generative AI through Karan Malhotra's WorldSim terminal environments, Rob Haisfield's WebSim generative brow…
49 statements 12 featured 11 assessments 2 contradicted
🎙LATENT SPACE
1h 2m High Agency Pydantic over VC Backed Frameworks — with Jason Liu of Instructor
In this episode of Latent Space, Jason Liu, creator of Instructor, joins Alessio and Swix to discuss the mechanics of structured LLM outputs, the architectural advantages of pragmatic engine…
27 statements 8 featured 1 assessments
🎙LATENT SPACE
1h 5m Breaking down the OG GPT Paper by Alec Radford
Machine learning engineer Amget presents a comprehensive technical breakdown of OpenAI's seminal 2018 GPT-1 paper, exploring how transformer-based generative pre-training revolutionized natu…
0 statements
🎙LATENT SPACE
1h 5m Supervise the Process of AI Research — with Jungwon Byun and Andreas Stuhlmüller of Elicit
Elicit co-founders Andreas Stuhlmüller and Jungwon Byun join the Latent Space Podcast to discuss building an AI research assistant for scientific literature, detailing their journey from ali…
27 statements 8 featured 1 assessments
🎙LATENT SPACE
58m Personal AI Meetup - Bee, BasedHardware, LangChain LangFriend, Deepgram EmilyAI
This meetup covers the tripartite foundation of personal AI companions, showcasing innovations in low-latency voice pipelines, open-source wearable hardware for continuous life logging, and …
9 statements 3 featured 1 assessments
🎙LATENT SPACE
49m Why Google failed to make GPT-3 -- with David Luan of Adept
Adept founder and former OpenAI VP David Luan discusses the organizational history of frontier AI scaling, why Google missed building GPT-3, and Adept's technical vision for multimodal auton…
16 statements 5 featured 1 assessments
🎙LATENT SPACE
54m A Comprehensive Overview of Large Language Models - Latent Space Paper Club
In this Latent Space Paper Club presentation, Brian delivers a comprehensive technical overview of large language models, tracing the historical evolution of attention mechanisms and Transfo…
0 statements
🎙LATENT SPACE
58m Making Transformers Sing - with Mikey Shulman of Suno
In this episode of the Live in Space podcast, co-hosts Alessio Fanelli and Shawn Wang interview Suno co-founder Mikey Shulman about the technical architecture, product design, and creative v…
23 statements 9 featured 3 assessments 1 contradicted
🎙LATENT SPACE
1h 37m Open Source AI is AI we can Trust — with Soumith Chintala of Meta AI
In this episode of the Latent Space Podcast, PyTorch creator and Meta AI Fellow Soumith Chintala explores the architectural evolution of deep learning frameworks, the strategic imperative fo…
29 statements 9 featured 4 assessments 1 contradicted
🎙LATENT SPACE
1h 21m A Brief History of the Open Source AI Hacker - with Ben Firshman of Replicate
In this episode of the Latent Space podcast, Replicate co-founder and CEO Ben Firshman explores the evolution of machine learning infrastructure, open-source developer tooling, and the emerg…
16 statements 7 featured 2 assessments 1 contradicted
🎙LATENT SPACE
1h 8m Truly Serverless Infra for AI Engineers - with Erik Bernhardsson of Modal
In this episode of the Latent Space Podcast, Modal founder and CEO Erik Bernhardsson discusses modernizing cloud infrastructure for AI and data engineering, highlighting custom container run…
31 statements 7 featured 4 assessments
🎙LATENT SPACE
1h 15m Building an open AI company - with Ce and Vipul of Together AI
Together AI co-founders Vipul Ved Prakash and Ce Zhang join the Latent Space podcast to discuss their open-source platform, disaggregated cloud infrastructure, and multi-dimensional inferenc…
26 statements 7 featured 5 assessments
🎙LATENT SPACE
1h 6m The State of AI in production — with David Hsu of Retool
Retool founder and CEO David Hsu joins the Latent Space podcast to discuss the evolution of Retool, contrarian principles of startup building, and the practical implementation of AI across m…
28 statements 8 featured 1 assessments
🎙LATENT SPACE
1h 20m The Four Wars of the AI Stack - Dec 2023 Recap
In this episode of the Latent Space podcast, co-hosts Alessio and Swix introduce 'The Four Wars of the AI Stack' framework to dissect the most critical battlegrounds shaping artificial intel…
0 statements
🎙LATENT SPACE
1h 35m The Origin and Future of RLHF: the secret ingredient for ChatGPT - with Nathan Lambert
In this episode of the Latent Space Podcast, AI researcher Dr. Nathan Lambert delivers a comprehensive masterclass on Reinforcement Learning from Human Feedback (RLHF), exploring its mathema…
57 statements 15 featured 15 assessments 1 contradicted
🎙LATENT SPACE
1h 27m The Accidental AI Canvas - with Steve Ruiz of tldraw
In this episode of Latent Space, tldraw founder Steve Ruiz recounts his journey from fine arts to software engineering and explains how his open-source infinite canvas engine became an essen…
20 statements 6 featured 3 assessments
🎙LATENT SPACE
1h 9m The AI-First Graphics Editor - with Suhail Doshi of Playground AI
Suhail Doshi, founder of Mixpanel and Playground AI, discusses the engineering and interface principles behind generative image models, detailing the training of Playground v2, the need for …
20 statements 5 featured 1 assessments 1 contradicted