In this episode, Zoe Weil, co-founder and CEO of @sequen_ai, joins us to discuss how ranking systems can use live behavioral signals to optimize for purchases rather than clicks or search relevance alone. Zoe explains why approaches developed at data-rich social platforms don’t necessarily translate to retailers and how Sequen uses scroll speed, hover duration, and other in-session interactions to infer shopper intent without relying on user identity. We explore the company’s large event models and its architecture of specialized models that transform streaming events into features for downstream rankers. Zoe describes the infrastructure required to serve dynamic rankings within milliseconds, why the team chose Rust for its backend, and how evaluation harnesses connect model improvements to business outcomes. We also discuss the tradeoffs between advertising and organic revenue, continuous model improvement, and the boundary between useful personalization and surveillance. 🗒️ Full show notes: twimlai.com/go/780. 📖 CHAPTERS =============================== 00:00 - Introduction 02:56 - Sequin’s Ranking Intelligence Layer 07:26 - Difference with Search and Relevance 13:20 - Session Live Signaling 15:57 - Surveillance Capitalism 18:45 - Building a Low-Latency Stack in Rust 20:52 - Model Architecture 26:47 - Reranking Products and Surfacing Complementary Items 33:03 - Balancing Organic and Ad Revenue 34:56 - Learning from Implicit Feedback 36:42 - Infrastructure 39:34 - Recursive Stack and Evaluation Harnesses 42:31 - Recursive Ranking Intelligence 44:12 - Large Event Models 46:56 - Time-Series Modeling 48:49 - Complexity of Scaling Real-Time Feature Transformations 49:46 - Future Directions

Get ready to join our Generative AI study group in one hour! Make sure you’re registered at twimlai.com/community/ to participate in the informative discussion, demonstrate personal projects, showcase cutting-edge tools and apps, and much more! See you there!

Be part of tomorrow's session of our Generative AI study group. Join us by registering at twimlai.com/community/ and explore decoder-style transformers, share insights on latency and throughput optimization, exchange alignment training strategies, and more. Our study group is open to everyone, so feel free to invite your peers!

Join the conversation in our Generative AI study group in one hour! Share model compression techniques, explore open-weight reasoning models, dig into embedding architectures, and more! Meet you in an hour!

Join our next Generative AI study group session tomorrow at 8 am PT and dig into long-context scaling techniques, explore linear attention hybrids, discuss orchestration frameworks for multi-agent workflows, and more! Register at twimlai.com/community/.

Join our interactive Generative AI study group every Friday at 8 am PT where we chat about the latest news in AI, explore new ideas, exchange perspectives and recommendations, and more! Registration is available at twimlai.com/community/ and don’t forget to join the #generative-ai channel in our Slack community.

In this episode, @CompleteSkeptic, co-founder and CEO of @typesafeai, joins us to discuss Jev, TypeSafe’s recently released model for bringing fast, reliable intelligence directly into software. We explore the idea of “machine-native intelligence” and why Diogo believes models optimized for generating text are poorly suited to many of the decisions required for real-world automation. He explains how Jev differs from traditional classifiers and LLM-based approaches, the role of reinforcement learning from calibrated decisions (RLCD), and why calibration and reliability are central to making AI useful as a software primitive. We also discuss the relationship between models and code, why Diogo believes AI systems should become more engineered rather than relying on a single model to do everything, and how Jev-like models could reshape agents, tool use, and the architecture of AI-powered software. 🗒️ Full show notes: twimlai.com/go/779. 📖 CHAPTERS =============================== 00:00 - Introduction 01:58 - Hot Takes 07:42 - The Bitterest Lesson 12:25 - Diogo’s Career Background 13:58 - Sci-Fi Software Demos vs. Easy Unsolved Problems 21:13 - Jev 22:37 - How Jev Differs from Classifiers and Small Models 32:20 - Intelligence and RLCD 35:57 - Jev as an Intelligence API 38:47 - RLCD Post-training 46:01 - RLCD and Calibrated Decisions 54:06 - Uncertainty Estimation 57:43 - Generalization 01:08:25 - Models, Code, and Harness 01:16:28 - Jev and the Future of AI Agents 01:22:21 - Rigid Code Workflows vs. Dynamic Agent Loops 01:27:05 - What’s Next

Join today’s session of our Generative AI study group in 1 hour! Sign up at twimlai.com/community/ and discuss memory bottlenecks, exchange LLM generation strategies, explore coding agent integrations, and more! See you there!

AI systems have gone from struggling with grade-school math to helping solve research problems that have resisted mathematicians for decades, including Navier-Stokes. In this episode, @GregHBurnham, who leads AI capabilities research at @EpochAIResearch, joins us to examine what that progress says about where AI is going. We look at how these systems are solving hard math problems, how much they rely on persistence and prior human work, and whether they are starting to produce genuinely new ideas. We also discuss how to measure progress as traditional benchmarks become less useful, why capability gains appear surprisingly steady across model generations, and where models still struggle with open-ended work, learning from experience, and identifying promising new research directions. 🗒️ Full show notes: twimlai.com/go/778. 📖 CHAPTERS =============================== 00:00 - Introduction 01:52 - AI Capabilities Research 03:42 - Is AI Progress Slowing Down? 06:38 - Correlation and Linear Trend of Benchmarks 13:55 - Drivers of AI Progress 19:52 - Evolution of AI Math Capabilities 28:27 - How AI Solves Hard Math Problems 17:05 - Detecting AI Research Breakthroughs 30:24 - AI’s Advantages: Knowledge and Persistence 36:16 - Defining New Ideas and Measuring AI Creativity 44:00 - FrontierMath 46:23 - Benchmarking When the Answers Are Unknown 47:54 - Evaluating Partial Progress on Problems 49:45 - Can AI Do Your Job? 50:23 - Open-Ended Work and Learning on the Fly 54:12 - Models Versus Harnesses 01:00:39 - In-Context On-the-Fly Learning 01:01:40 - Big Questions Shaping AI’s Progress and Trajectory

Only 1 hour left before our Generative AI study group starts! ⏳ Don’t miss out on insightful and thought-provoking conversations about AI agents, multimodality, test-time scaling, and more! See you in an hour! #llms #generativeai #studygroup #TWIMLMeetup #TWIML #machinelearning #ml #ai #artificialintelligence

Head over to twimlai.com/community/ to register for tomorrow's session of our Generative AI study group. Explore image-text encoders, exchange latent reasoning approaches, dig into self-verification in reinforcement learning, and more! Meetup is at 8 am PT.

Be part of our TWIML community and join our Generative AI study group every Friday at 8 am PT. Dig into visual grounding techniques, explore Mixture-of-Modality models, discuss agent governance, and more! Feel free to invite a friend and register at twimlai.com/community. See you there!

1 hour left before this week’s meetup of our Generative AI study group starts! If you haven’t registered yet, head over now to twimlai.com/community/ and don’t forget to join the #generative-ai Slack channel. See you there!

Visit twimlai.com/community/ to register for tomorrow's Generative AI study group at 8 am PT. Discuss distributed inference pipelines, exchange ideas on reasoning-time scaling techniques, explore multimodal transformer models, and more.

Voice AI has gotten remarkably good, but natural conversation remains a high bar. Small delays, awkward interruptions, or the wrong tone can quickly break the illusion, and adding vision and visual presence only raises the stakes. In this episode, @smolix, co-founder and CEO of @boson_ai, explores the path from today’s voice agents to audiovisual agents and AI avatars. We discuss the technical tradeoffs behind real-time voice, including audio tokenization, latency, model size, and inference cost, as well as what changes when these systems can both see and be seen. We also explore the role of emotional intelligence in AI, how agents can learn from human interactions, and what it will take to move beyond impressive demos toward interactions that actually feel natural. 🗒️ Full show notes: twimlai.com/go/777. 📖 CHAPTERS =============================== 00:00 - Introduction 01:36 - Challenges of Voice AI in Real-World Environments 06:12 - Why Voice Systems Are a Whole-System Challenge 12:50 - Trade-Off Between Size, Compute, and Affordability 13:53 - Boson AI’s Path to Audio 19:59 - Training Dataset 23:49 - Dealing with Noisy Data 28:41 - Model Training Approach 33:14 - Architectural Decisions: End-to-End vs. Multi-Threaded Systems 39:26 - MCP, Context Windows, and Inference Costs 41:04 - Agentic Systems for Voice 46:31 - Benchmarking 50:03 - Why EQ Matters for Voice and Video 53:43 - Recursive Self-Improvement and Personalization 01:00:03 - Brilliance, Interaction Improvement, and Wrap-Up

Don’t miss out and hop in an hour to meet us in our Generative AI study group! Ask questions, share your insights, learn alongside fellow AI practitioners, and more! See you there!

Join us for another engaging conversation around open-source distributed inference frameworks, LVLM spatial understanding, AI observability for agentic systems, and more in our Generative AI study group! Visit twimlai.com/community/ to register and we’ll see you tomorrow at 8 am PT.

Join our weekly Generative AI study group every Friday at 8 am PT! Dig into geospatial reasoning, share strategies for patch-text alignment, discuss KV cache routing, and more! For registration, head over to twimlai.com/community/ to sign up.

As reasoning models consume more tokens and AI systems become more expensive to run, understanding what those tokens actually buy is becoming increasingly important. In this episode, @Stanford professor and Big Spin co-founder @ChrisGPotts joins us to discuss AI tokenomics and his research into “tokenflation”—the possibility that token usage is growing faster than the measurable value those tokens produce. We explore how to measure the return on AI spending, why benchmarks alone provide an incomplete picture of model progress, and what inference-time scaling means for the economics of increasingly capable models. Chris also explains why expert AI users tend to get better results by challenging and iterating with models, how AI fluency affects outcomes, and why more efficient architectures could change the underlying economics. We also discuss DSPy, interpretability, the limits of today’s transformer architectures, and where Chris sees opportunities for more fundamental innovation in AI. 🗒️ Full show notes: twimlai.com/go/776. 📖 CHAPTERS =============================== 00:00 - Introduction 05:38 - Linguistics in the Age of Language Models 09:26 - Scale Limitations in NLP Research 12:54 - Challenging the Bitter Lesson Mindset 15:12 - Relationship Between Data, Mechanistic Interpretability, and Efficiency 17:03 - DSPy 21:32 - Prompt Optimization and Model Variability 24:35 - Tokenomics and the Rising Cost of AI 28:13 - Measuring Token Purchasing Power with a CPI 32:18 - Inference-Time Scaling 35:36 - Defining Value Across Different AI Tasks 38:33 - AI Value Creation 40:38 - Predicting AI Costs 42:25 - Tokenflation 46:22 - AI Fluency 50:12 - Key Lessons of AI Fluency Work 54:44 - Future Directions

Get ready to join us in 1 hour for our Generative AI study group! ⏳ Engage in the AI news discussion, bring your questions, exchange ideas, share experiences, and learn together! Meet you there!