Proactive Agents for the Web with Devi Parikh - #756

Today, we're joined by Stefano Ermon, associate professor at Stanford University and CEO of Inception Labs to discuss diffusion language models. We dig into how diffusion approaches—traditionally used for images—are being adapted for text and code generation, the technical challenges of applying continuous methods to discrete token spaces, and how diffusion models compare to traditional autoregressive LLMs. Stefano introduces Mercury 2, a commercial-scale diffusion LLM that can generate multiple tokens simultaneously and achieve inference speeds 5-10x faster than small frontier models, paving the way for latency-sensitive applications like voice interactions and fast agentic loops. We also cover the open research challenges in diffusion LLM training, serving infrastructure requirements, and post-training for diffusion-based systems. Finally, Stefano shares his perspective on whether diffusion models can rival or surpass autoregressive LLMs at scale, the advantages for highly controllable generation, and what the future of multimodal diffusion models might look like. The complete show notes for this episode can be found at https://twimlai.com/go/764.

In this episode, Sid Pardeshi, co-founder and CTO of Blitzy, joins us to discuss building autonomous development systems able to deliver production-ready software at enterprise scale. Sid contrasts AI-assisted coding with end-to-end autonomy, arguing that “code is a commodity” and acceptance is the real metric—security, standards, tests, and maintainability included. We explore Blitzy’s hybrid graph-plus-vector approach, which grounds agents and combines semantic signals with keyword search to navigate large repositories efficiently. Sid breaks down context and agent engineering, how effective context windows have plateaued, and why dynamic agent personas, tool selection, and model-specific prompting matter at scale. He details their orchestration of large swarms of AI agents to collaboratively analyze codebases, plan tasks, and execute complex tasks in parallel. We also dig into why Agents.md and flat memories break down, storing feedback in the knowledge graph, and building real-world evals beyond leaderboards to choose the right model for each task. The complete show notes for this episode can be found at https://twimlai.com/go/763.

In this episode, Sebastian Raschka, independent LLM researcher and author, joins us to break down how the LLM landscape has changed over the past year and what is likely to matter most in 2026. We discuss the shift from raw model scaling to reasoning-focused post-training, inference-time techniques, and better tool integration. Sebastian explains why methods like self-consistency, self-refinement, and verifiable-reward reinforcement learning have become central to progress in domains like math and coding, and where those approaches still fall short. We also explore agentic workflows in practice, including where multi-agent systems add real value and where reliability constraints still dominate system design. The conversation covers architecture trends such as mixture-of-experts, attention efficiency strategies, and the practical impact of long-context models, alongside persistent challenges like continual learning. We close with Sebastian’s perspective on maintaining strong coding fundamentals in the age of AI assistants and a preview of his new book, Build A Reasoning Model (From Scratch). The complete show notes for this episode can be found at https://twimlai.com/go/762.

Today, we're joined by Yejin Choi, professor and senior fellow at Stanford University in the Computer Science Department and the Institute for Human-Centered AI (HAI). In this conversation, we explore Yejin’s recent work on making small language models reason more effectively. We discuss how high-quality, diverse data plays a central role in closing the intelligence gap between small and large models, and how combining synthetic data generation, imitation learning, and reinforcement learning can unlock stronger reasoning capabilities in smaller models. Yejin explains the risks of homogeneity in model outputs and mode collapse highlighted in her “Artificial Hivemind” paper, and its impacts on human creativity and knowledge. We also discuss her team's novel approaches, including reinforcement learning as a pre-training objective, where models are incentivized to “think” before predicting the next token, and "Prismatic Synthesis," a gradient-based method for generating diverse synthetic math data while filtering overrepresented examples. Additionally, we cover the societal implications of AI and the concept of pluralistic alignment—ensuring AI reflects the diverse norms and values of humanity. Finally, Yejin shares her mission to democratize AI beyond large organizations and offers her predictions for the coming year. The complete show notes for this episode can be found at https://twimlai.com/go/761.

Today, we're joined by Nikita Rudin, co-founder and CEO of Flexion Robotics to discuss the gap between current robotic capabilities and what’s required to deploy fully autonomous robots in the real world. Nikita explains how reinforcement learning and simulation have driven rapid progress in robot locomotion—and why locomotion is still far from “solved.” We dig into the sim2real gap, and how adding visual inputs introduces noise and significantly complicates sim-to-real transfer. We also explore the debate between end-to-end models and modular approaches, and why separating locomotion, planning, and semantics remains a pragmatic approach today. Nikita also introduces the concept of "real-to-sim", which uses real-world data to refine simulation parameters for higher fidelity training, discusses how reinforcement learning, imitation learning, and teleoperation data are combined to train robust policies for both quadruped and humanoid robots, and introduces Flexion's hierarchical approach that utilizes pre-trained Vision-Language Models (VLMs) for high-level task orchestration with Vision-Language-Action (VLA) models and low-level whole-body trackers. Finally, Nikita shares the behind-the-scenes in humanoid robot demos, his take on reinforcement learning in simulation versus the real world, the nuances of reward tuning, and offers practical advice for researchers and practitioners looking to get started in robotics today. The complete show notes for this episode can be found at https://twimlai.com/go/760.

Today, we're joined by Aakanksha Chowdhery, member of technical staff at Reflection, to explore the fundamental shifts required to build true agentic AI. While the industry has largely focused on post-training techniques to improve reasoning, Aakanksha draws on her experience leading pre-training efforts for Google’s PaLM and early Gemini models to argue that pre-training itself must be rethought to move beyond static benchmarks. We explore the limitations of next-token prediction for multi-step workflows and examine how attention mechanisms, loss objectives, and training data must evolve to support long-form reasoning and planning. Aakanksha shares insights on the difference between context retrieval and actual reasoning, the importance of "trajectory" training data, and why scaling remains essential for discovering emergent agentic capabilities like error recovery and dynamic tool learning. The complete show notes for this episode can be found at https://twimlai.com/go/759.

In this episode, we’re joined by Munawar Hayat, researcher at Qualcomm AI Research, to discuss a series of papers presented at NeurIPS 2025 focusing on multimodal and generative AI. We dive into the persistent challenge of object hallucination in Vision-Language Models (VLMs), why models often discard visual information in favor of pre-trained language priors, and how his team used attention-guided alignment to enforce better visual grounding. We also explore a novel approach to generalized contrastive learning designed to solve complex, composed retrieval tasks—such as searching via combined text and image queries—without increasing inference costs. Finally, we cover the difficulties generative models face when rendering multiple human subjects, and the new "MultiHuman Testbench" his team created to measure and mitigate issues like identity leakage and attribute blending. Throughout the discussion, we examine how these innovations align with the need for efficient, on-device AI deployment. The complete show notes for this episode can be found at https://twimlai.com/go/758.

In this episode, Zain Asgar, co-founder and CEO of Gimlet Labs, joins us to discuss the heterogeneous AI inference across diverse hardware. Zain argues that the current industry standard of running all AI workloads on high-end GPUs is unsustainable for agents, which consume significantly more tokens than traditional LLM applications. We explore Gimlet’s approach to heterogeneous inference, which involves disaggregating workloads across a mix of hardware—from H100s to older GPUs and CPUs—to optimize unit economics without sacrificing performance. We dive into their "three-layer cake" architecture: workload disaggregation, a compilation layer that maps models to specific hardware targets, and a novel system that uses LLMs to autonomously rewrite and optimize compute kernels. Finally, we discuss the complexities of networking in heterogeneous environments, the trade-offs between numerical precision and application accuracy, and the future of hardware-aware scheduling. The complete show notes for this episode can be found at https://twimlai.com/go/757.

Today, we're joined by Devi Parikh, co-founder and co-CEO of Yutori, to discuss browser use models and a future where we interact with the web through proactive, autonomous agents. We explore the technical challenges of creating reliable web agents, the advantages of visually-grounded models that operate on screenshots rather than the browser’s more brittle document object model, or DOM, and why this counterintuitive choice has proven far more robust and generalizable for handling complex web interfaces. Devi also shares insights into Yutori’s training pipeline, which has evolved from supervised fine-tuning to include rejection sampling and reinforcement learning. Finally, we discuss how Yutori’s “Scouts” agents orchestrate multiple tools and sub-agents to handle complex queries, the importance of background, "ambient" operation for these systems, and what the path looks like from simple monitoring to full task automation on the web. The complete show notes for this episode can be found at https://twimlai.com/go/756.

Today, we're joined by Robin Braun, VP of AI business development for hybrid cloud at HPE, and Luke Norris, co-founder and CEO of Kamiwaza, to discuss how AI systems can be used to automate complex workflows and unlock value from legacy enterprise data. Robin and Luke detail high-impact use cases from HPE and Kamiwaza’s collaboration on an “Agentic Smart City” project for Vail, Colorado, including remediation and automation of website accessibility for 508 compliance, digitization and understanding of deed restrictions, and combining contextual information with camera feeds for fire detection and risk assessment. Additionally, we discuss the role of private cloud infrastructure in overcoming challenges like cost, data privacy, and compliance. Robin and Luke also share their lessons learned, including the importance of fresh data, and the value of a "mud puddle by mud puddle" approach in achieving practical AI wins. The complete show notes for this episode can be found at https://twimlai.com/go/755.