Best Intelligent Speculation Podcasts (2025)

1
166 // Blade Runner Behind-The-Scenes with Assistant Director Vicki Rhodes 58:14

Play Pause

3d ago58:14

58:14

In this ultra-special episode, Jaime and Patrick sit down with Vicki Rhodes, assistant director on Blade Runner, to uncover the incredible behind-the-scenes stories of this cinematic landmark. From all-night shoots to ostrich wrangling, Vicki shares a perspective few have ever heard, a journey into the making of a film that continues to define and …

1
Google I/O 2025 Special Edition - #733 26:21

7d ago26:21

26:21

Today, I’m excited to share a special crossover edition of the podcast recorded live from Google I/O 2025! In this episode, I join Shawn Wang aka Swyx from the Latent Space Podcast, to interview Logan Kilpatrick and Shrestha Basu Mallick, PMs at Google DeepMind working on AI Studio and the Gemini API, along with Kwindla Kramer, CEO of Daily and cre…

1
RAG Risks: Why Retrieval-Augmented LLMs are Not Safer with Sebastian Gehrmann - #732 57:09

14d ago57:09

57:09

Today, we're joined by Sebastian Gehrmann, head of responsible AI in the Office of the CTO at Bloomberg, to discuss AI safety in retrieval-augmented generation (RAG) systems and generative AI in high-stakes domains like financial services. We explore how RAG, contrary to some expectations, can inadvertently degrade model safety. We cover examples o…

1
From Prompts to Policies: How RL Builds Better AI Agents with Mahesh Sathiamoorthy - #731 1:01:25

22d ago1:01:25

1:01:25

Today, we're joined by Mahesh Sathiamoorthy, co-founder and CEO of Bespoke Labs, to discuss how reinforcement learning (RL) is reshaping the way we build custom agents on top of foundation models. Mahesh highlights the crucial role of data curation, evaluation, and error analysis in model performance, and explains why RL offers a more robust altern…

1
How OpenAI Builds AI Agents That Think and Act with Josh Tobin - #730 1:07:27

29d ago1:07:27

1:07:27

Today, we're joined by Josh Tobin, member of technical staff at OpenAI, to discuss the company’s approach to building AI agents. We cover OpenAI's three agentic offerings—Deep Research for comprehensive web research, Operator for website navigation, and Codex CLI for local code execution. We explore OpenAI’s shift from simple LLM workflows to reaso…

1
165 // Off-World: The Costumes of Blade Runner 54:34

1M ago54:34

54:34

We're proud to announce our inaugural episode of our exclusive Patreon series, Off-World. In this episode, Patrick and Jaime are joined by patron and listener David Hanbury as we go through the costumes of the first Blade Runner film. What makes the designs of the costumes in Ridley Scott's seminal film so iconic? Which costume is your favorite? Do…

1
CTIBench: Evaluating LLMs in Cyber Threat Intelligence with Nidhi Rastogi - #729 56:18

1M ago56:18

56:18

Today, we're joined by Nidhi Rastogi, assistant professor at Rochester Institute of Technology to discuss Cyber Threat Intelligence (CTI), focusing on her recent project CTIBench—a benchmark for evaluating LLMs on real-world CTI tasks. Nidhi explains the evolution of AI in cybersecurity, from rule-based systems to LLMs that accelerate analysis by p…

1
Generative Benchmarking with Kelly Hong - #728 54:17

1M ago54:17

54:17

In this episode, Kelly Hong, a researcher at Chroma, joins us to discuss "Generative Benchmarking," a novel approach to evaluating retrieval systems, like RAG applications, using synthetic data. Kelly explains how traditional benchmarks like MTEB fail to represent real-world query patterns and how embedding models that perform well on public benchm…

1
Exploring the Biology of LLMs with Circuit Tracing with Emmanuel Ameisen - #727 1:34:06

2M ago1:34:06

1:34:06

In this episode, Emmanuel Ameisen, a research engineer at Anthropic, returns to discuss two recent papers: "Circuit Tracing: Revealing Language Model Computational Graphs" and "On the Biology of a Large Language Model." Emmanuel explains how his team developed mechanistic interpretability methods to understand the internal workings of Claude by rep…

1
Teaching LLMs to Self-Reflect with Reinforcement Learning with Maohao Shen - #726 51:45

2M ago51:45

51:45

Today, we're joined by Maohao Shen, PhD student at MIT to discuss his paper, “Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search.” We dig into how Satori leverages reinforcement learning to improve language model reasoning—enabling model self-reflection, self-correction, and exploration of a…

1
164 // Blade Runner Ties: Devs 44:13

2M ago44:13

44:13

Alex Garland's Devs is a heady sci-fi series that asks some Big Questions about the nature of reality, the things that make us human, and the inescapable reality of loss—so if you love Blade Runner, you'll likely love this show! In today's installment of our ongoing Blade Runner Ties series, Jaime and Patrick look at Devs through the lens of Blade …

1
Waymo's Foundation Model for Autonomous Driving with Drago Anguelov - #725 1:09:07

2M ago1:09:07

1:09:07

Today, we're joined by Drago Anguelov, head of AI foundations at Waymo, for a deep dive into the role of foundation models in autonomous driving. Drago shares how Waymo is leveraging large-scale machine learning, including vision-language models and generative AI techniques to improve perception, planning, and simulation for its self-driving vehicl…

1
Random number generator [FREE bonus episode] 1:50

2M ago1:50

1:50

Join the show's Patreon at https://www.patreon.com/programaudioseries FOR FREE to listen to this episode. You'll also get monthly updates, quizzes, and access to a community of fellow Program listeners at no cost. You can upgrade to a paid membership at any time to get bonus episodes and ad-free listening for as little as $3/mo. Get bonus content o…

1
Dynamic Token Merging for Efficient Byte-level Language Models with Julie Kallini - #724 50:32

2M ago50:32

50:32

Today, we're joined by Julie Kallini, PhD student at Stanford University to discuss her recent papers, “MrT5: Dynamic Token Merging for Efficient Byte-level Language Models” and “Mission: Impossible Language Models.” For the MrT5 paper, we explore the importance and failings of tokenization in large language models—including inefficient compression…

1
Scaling Up Test-Time Compute with Latent Reasoning with Jonas Geiping - #723 58:38

3M ago58:38

58:38

Today, we're joined by Jonas Geiping, research group leader at Ellis Institute and the Max Planck Institute for Intelligent Systems to discuss his recent paper, “Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach.” This paper proposes a novel language model architecture which uses recurrent depth to enable “thinking in l…

1
Imagine while Reasoning in Space: Multimodal Visualization-of-Thought with Chengzu Li - #722 42:11

3M ago42:11

42:11

Today, we're joined by Chengzu Li, PhD student at the University of Cambridge to discuss his recent paper, “Imagine while Reasoning in Space: Multimodal Visualization-of-Thought.” We explore the motivations behind MVoT, its connection to prior work like TopViewRS, and its relation to cognitive science principles such as dual coding theory. We dig i…

1
Inside s1: An o1-Style Reasoning Model That Cost Under $50 to Train with Niklas Muennighoff - #721 49:29

3M ago49:29

49:29

Today, we're joined by Niklas Muennighoff, a PhD student at Stanford University, to discuss his paper, “S1: Simple Test-Time Scaling.” We explore the motivations behind S1, as well as how it compares to OpenAI's O1 and DeepSeek's R1 models. We dig into the different approaches to test-time scaling, including parallel and sequential scaling, as well…

1
163 // Blade Runner Ties: HER 1:01:53

3M ago1:01:53

1:01:53

In the second installment of our Blade Runner Ties series, Patrick, Peter, and Jaime dive into Spike Jonze’s Her (2013). Set in a near-future world, the film tells the story of Theodore, a man who falls in love with his operating system—voiced beautifully by Scarlett Johansson. While there are undeniable connections to Blade Runner 2049 in the char…

1
Accelerating AI Training and Inference with AWS Trainium2 with Ron Diamant - #720 1:07:05

3M ago1:07:05

1:07:05

Today, we're joined by Ron Diamant, chief architect for Trainium at Amazon Web Services, to discuss hardware acceleration for generative AI and the design and role of the recently released Trainium2 chip. We explore the architectural differences between Trainium and GPUs, highlighting its systolic array-based compute design, and how it balances per…

1
π0: A Foundation Model for Robotics with Sergey Levine - #719 52:30

4M ago52:30

52:30

Today, we're joined by Sergey Levine, associate professor at UC Berkeley and co-founder of Physical Intelligence, to discuss π0 (pi-zero), a general-purpose robotic foundation model. We dig into the model architecture, which pairs a vision language model (VLM) with a diffusion-based action expert, and the model training "recipe," emphasizing the ro…

1
AI Trends 2025: AI Agents and Multi-Agent Systems with Victor Dibia - #718 1:44:59

4M ago1:44:59

1:44:59

Today we’re joined by Victor Dibia, principal research software engineer at Microsoft Research, to explore the key trends and advancements in AI agents and multi-agent systems shaping 2025 and beyond. In this episode, we discuss the unique abilities that set AI agents apart from traditional software systems–reasoning, acting, communicating, and ada…

1
Speculative Decoding and Efficient LLM Inference with Chris Lott - #717 1:16:30

4M ago1:16:30

1:16:30

Today, we're joined by Chris Lott, senior director of engineering at Qualcomm AI Research to discuss accelerating large language model inference. We explore the challenges presented by the LLM encoding and decoding (aka generation) and how these interact with various hardware constraints such as FLOPS, memory footprint and memory bandwidth to limit…

1
162 // Blade Runner Ties: Ex Machina 42:10

4M ago42:10

42:10

Jaime and Patrick kick off a new series exploring the themes of Blade Runner through the lens of other films. In this first episode, they dive deep into Alex Garland’s Ex Machina, examining its parallels to Blade Runner and where the two films diverge. This episode is thought-provoking and rich with discussion—everything this podcast is at its core…

1
Ensuring Privacy for Any LLM with Patricia Thaine - #716 51:33

4M ago51:33

51:33

Today, we're joined by Patricia Thaine, co-founder and CEO of Private AI to discuss techniques for ensuring privacy, data minimization, and compliance when using 3rd-party large language models (LLMs) and other AI services. We explore the risks of data leakage from LLMs and embeddings, the complexities of identifying and redacting personal informat…

1
AI Engineering Pitfalls with Chip Huyen - #715 57:37

4M ago57:37

57:37

Today, we're joined by Chip Huyen, independent researcher and writer to discuss her new book, “AI Engineering.” We dig into the definition of AI engineering, its key differences from traditional machine learning engineering, the common pitfalls encountered in engineering AI systems, and strategies to overcome them. We also explore how Chip defines …

1
Evolving MLOps Platforms for Generative AI and Agents with Abhijit Bose - #714 58:08

5M ago58:08

58:08

Today, we're joined by Abhijit Bose, head of enterprise AI and ML platforms at Capital One to discuss the evolution of the company’s approach and insights on Generative AI and platform best practices. In this episode, we dig into the company’s platform-centric approach to AI, and how they’ve been evolving their existing MLOps and data platforms to …

1
161 // 2024 in Review 44:41

5M ago44:41

44:41

In this recap of 2024, Patrick and Jaime sit down to review highlights of the year, the wrapping of Blade Runner 2099, notable discussions and more. We have a lot in store for 2025. With what we hope will be a fantastic new live action series from Alcon and Amazon, there will be a lot to discuss in the coming months. Thank you for listening to the …

1
Free trial period (3rd instalment) 28:52

6M ago28:52

28:52

Nine strangers are taken hostage. Some will be exchanged in a prisoner swap. Others will be sacrificed. They need to decide who will be who. This is the third out of three instalments. Listen to the full episode ad-free by supporting the show on Patreon at https://www.patreon.com/programaudioseries Visit programaudioseries.com/35-free-trial-period-…

1
Why Agents Are Stupid & What We Can Do About It with Dan Jeffries - #713 1:08:49

6M ago1:08:49

1:08:49

Today, we're joined by Dan Jeffries, founder and CEO of Kentauros AI to discuss the challenges currently faced by those developing advanced AI agents. We dig into how Dan defines agents and distinguishes them from other similar uses of LLM, explore various use cases for them, and dig into ways to create smarter agentic systems. Dan shared his “big …

1
Free trial period (2nd instalment) 19:50

6M ago19:50

19:50

Nine strangers are taken hostage. Some will be exchanged in a prisoner swap. Others will be sacrificed. They need to decide who will be who. This is the second out of three instalments. Listen to the full episode ad-free by supporting the show on Patreon at https://www.patreon.com/programaudioseries Visit programaudioseries.com/35-free-trial-period…

1
Automated Reasoning to Prevent LLM Hallucination with Byron Cook - #712 56:48

6M ago56:48

56:48

Today, we're joined by Byron Cook, VP and distinguished scientist in the Automated Reasoning Group at AWS to dig into the underlying technology behind the newly announced Automated Reasoning Checks feature of Amazon Bedrock Guardrails. Automated Reasoning Checks uses mathematical proofs to help LLM users safeguard against hallucinations. We explore…

1
Free trial period (1st instalment) 21:31

6M ago21:31

21:31

Nine strangers are taken hostage. Some will be exchanged in a prisoner swap. Others will be sacrificed. They need to decide who will be who. This is the first out of three instalments. Listen to the full episode ad-free by supporting the show on Patreon at https://www.patreon.com/programaudioseries Visit programaudioseries.com/35-free-trial-period-…

1
AI at the Edge: Qualcomm AI Research at NeurIPS 2024 with Arash Behboodi - #711 54:47

6M ago54:47

54:47

Today, we're joined by Arash Behboodi, director of engineering at Qualcomm AI Research to discuss the papers and workshops Qualcomm will be presenting at this year’s NeurIPS conference. We dig into the challenges and opportunities presented by differentiable simulation in wireless systems, the sciences, and beyond. We also explore recent work that …

1
160 // Home Again Home Again: Interviewing Blade Runner Actor Kevin Thompson 57:41

6M ago57:41

57:41

In this exclusive interview with Shoulder of Orion: The Blade Runner Podcast, Patrick & Jaime are joined by Blade Runner actor Kevin Thompson. Kevin most famously played J.F. Sebastian's toy bear in Blade Runner as well as a myriad of other roles. Kevin takes us through his time on Blade Runner, his relationship with Ridley Scott, Rutger Hauer and …

1
AI for Network Management with Shirley Wu - #710 53:44

7M ago53:44

53:44

Today, we're joined by Shirley Wu, senior director of software engineering at Juniper Networks to discuss how machine learning and artificial intelligence are transforming network management. We explore various use cases where AI and ML are applied to enhance the quality, performance, and efficiency of networks across Juniper’s customers, including…

1
Why Your RAG System Is Broken, and How to Fix It with Jason Liu - #709 58:03

7M ago58:03

58:03

Today, we're joined by Jason Liu, freelance AI consultant, advisor, and creator of the Instructor library to discuss all things retrieval-augmented generation (RAG). We dig into the tactical and strategic challenges companies face with their RAG system, the different signs Jason looks for to identify looming problems, the issues he most commonly en…

1
An Agentic Mixture of Experts for DevOps with Sunil Mallya - #708 1:15:09

7M ago1:15:09

1:15:09

Today we're joined by Sunil Mallya, CTO and co-founder of Flip AI. We discuss Flip’s incident debugging system for DevOps, which was built using a custom mixture of experts (MoE) large language model (LLM) trained on a novel "CoMELT" observability dataset which combines traditional MELT data—metrics, events, logs, and traces—with code to efficientl…

1
Stories of Denial, Despair, and Devotion 16:19

7M ago16:19

16:19

A collection of didactic micro stories from the far future. Presented with Decoders' commentary. Visit programaudioseries.com/34-stories-of-denial-despair-and-devotion for full transcript, credits, and list of references. Join the show's Patreon at https://www.patreon.com/programaudioseries FOR FREE and get monthly updates, quizzes, and access to a…