Latent Space - Videos
Back to Channel
Dev Fixes Bug with Just One Prompt
Coding is just the start. One dev fixed a Cloudflare bug, closed a GitHub issue, and deployed—all inside VS Code using Klein. Natural language + MCP = next-level automation.
Klein’s Origin Story: Built in 10 Days
Klein was born just 10 days after Claude 3.5 dropped. Hear how one dev saw the future of agentic coding and built a tool that outpaced Copilot and Cursor in reasoning and long-context understanding
OpenCode: Claude Code but Open Source, with Any Model, and frontier TUI - with Dax Reed (@thdxr)
https://github.com/sst/opencode (https://github.com/sst/opencode#how-is-this-different-than-claude-code) We caught up with Dax of OpenCode and SST about his virally popular project, and his primary...
Why AI Writes BAD Code 🚫
The truth about AI code generation! Why current AI models write terrible defensive code and what this reveals about the limitations of reinforcement learning in programming
Plan & Act: How Devs Work with AI
Plan and Act"—Klein’s unique two-mode system that lets you guide your coding agent like a pro. Learn how devs use plan mode to brainstorm and act mode to execute, all while sipping coffee.
Klein The Agent That Codes for You
What is Klein? Discover how this open-source coding agent transforms your entire developer workflow. From VS Code to the browser, Klein takes over and gets the job done—so you can sit back and relax
⚡️Mercury: Ultra-Fast Diffusion LLMs — Estefano Ermon, CEO Inception Labs
https://arxiv.org/abs/2506.17298 We present Mercury, a new generation of commercial-scale large language models (LLMs) based on diffusion. These models are parameterized via the Transformer archit...
⚡️Composio: 10,000+ tools that evolve for Agents — Karan Vaidya and Soham Ganatra
We met with Karan and Soham, cofounders of Composio, to talk over their Series A announcement! https://composio.dev/blog/series-a AI agents today don't learn from experience. You can engineer prom...
AI Influencers Beauty Over Brain
The RLVR Revolution — with Nathan Lambert (AI2, Interconnects.ai)
00:00:00 Welcome and Guest Introduction 00:01:18 Tulu, OVR, and the RLVR Journey 00:03:40 Industry Approaches to Post-Training and Preference Data 00:06:08 Understanding RLVR and Its Impact 00:06:1...
Warp's Local Sandbox Advantage
Using RFT to Build Clinical Superintelligence
Brendan Fortuner, Head of Eng at Ambience AI, was one of the OpenAI RFT launch partner. Today we talk about how they used reinforcement fine tuning to help doctors spend less time taking notes in m...
🕰️ The Oral History of Windsurf (ft. Varun Mohan, Scott Wu, Jeff Wang, Kevin Hou, Anshul R)
*This is a recap episode that ends with a short fresh interview on the future of Windsurf + Cognition with Jeff Wang and Scott Wu at the end.* As the story of Windsurf as an independent company has...
Warp's Universal Input Auto Detecting Commands
⚡️Math Olympiad gold medalist explains OpenAI and Google DeepMind IMO Gold Performances
Math Olympiad gold medalist explains OpenAI and Google DeepMind IMO Gold Performances https://frontiermath-symposium.epoch.ai/ https://epoch.ai/frontiermath/tier-4 https://news.smol.ai/issue...
AI is Eating Search
ChatGPT handles 2.5B prompts/day and is on track to match Google's daily searches by end of 2026. AI agents don't browse like us—they crave queryable, chunkable data for tools like ChatGPT & Perple...
Beyond AlphaGo
The hidden flaws of minimax training, the myth of generalizable strategy, and designing AI for real-world complexity
AlphaZero vs. ChatGPT
What Strategic Games Taught Us About AI — and Why It Doesn’t Translate to Language
AI Agents That Pay for Their Own Tools? The Future of MCP Monetization
Wild vision from Cline: agents that pay themselves for MCP tools 🤯 Instead of juggling API keys for every service, imagine a unified payment layer where AI agents automatically handle transactio...
The Gathering Breaks AI
Noam explains why Magic: The Gathering is so hard for AI—and what it teaches us about hidden information and reasoning in games
Does AI Need Intuition First?
From Pokémon battles to chess, Noam dives into how much intuition (System 1) an AI needs before it can think deeply (System 2) about a game.
Arc Prize's Greg Kamradt: "Reality is the Ultimate Eval Engine"
Greg Kamradt from Arc Prize drops a profound insight: "Reality is the ultimate eval engine" 🎯 His take on AGI benchmarks is fascinating - Arc Prize doesn't aim for PhD-level impossible problems....
Why AI Models Aren't Actually Intelligent (Yet) - The Physics Problem
Humans RL on physics - the ultimate eval engine. But when AI uses manufactured environments, we're just injecting developer intelligence into the system. The real test? Interactive benchmarks wh...
Why AI Inference Will Never Be Profitable (Cline founder Explains)
🔥 HOT TAKE: "Inference isn't a business" - Cline founder drops truth bombs Key insights from this clip: • Model costs trending toward zero (Gemini going free?) • Price transparency ) obfuscati...
Why C++ Sucks for AI Programming
Why build a new programming language when everyone already uses Python and C++? Chris Lattner explains the origin of Mojo and why C++ "sucks" for AI workloads.
Interactive Data Selection: Marimo vs Jupyter - Game Changer for Data Scientists
The difference between Jupyter and Marimo is wild 🤯 With Marimo, you can literally drag-select data points in a scatter plot and instantly see the underlying images. No complex setup, no painful...
"Scaffolds Are All Going To Die" — Noam Brown's Spicy AGI Prediction
Noam Brown just dropped a spicy take: "Scaffolds are all going to die" 🔥 While most AI engineers are building complex agent frameworks with memory systems, compression layers, and elaborate harn...
Python Speed Without the Pain
Mojo can replace Python for performance-critical code—without bindings or rewrites. Chris shares how Mojo makes Rust-level speed accessible to Python devs.
Engineering With AI Teammates
Deep diving into next-gen development environments, tool-calling evolution, and how Zed is pioneering the integration of LLMs as first-class collaborators
Rust-Powered Editing at 120fps
Inside Zed's architecture: game-engine rendering, CRDTs for collaboration, and why web tech couldn’t go the distance
From Code Editors to Code Environments
The evolution of developer tools through AI curation, multi-buffer workflows, and human-centered agent design
Don’t Reinvent—Remix Smart Ideas
Chris Lattner explains how Modular decides what to build vs. what to adopt—and why composability and clean architecture win every time.
⚡️ARC-AGI-3: The Interactive Reasoning Benchmark
Greg Kamradt, President @ ARC Prize Foundation, joins us for a sneak preview of the ARC AGI 3 benchmark, how they are approaching game design for AI, and why this is a better benchmark than traditi...
Agentic Loop Font Kit & Context Gathering
DeepSeek’s GPU Hack Changed Everything
The DeepSeek team shocked the world by going deeper than CUDA—and open sourcing it. Chris Lattner reacts to the moment that woke up the AI industry
⚡️The Future of Notebooks - with Akshay Agrawal of Marimo
Akshay Agrawal joins us to talk about Marimo and their vision for the future of Python notebooks, and how it’s the perfect canvas for AI-driven data analysis. 0:00 Introduction 0:46 Overview of ...
CTO: "GPT-4 Changed Nothing Outside Silicon Valley" - The Reality Check AI Needs
CTO Andrew Struggs drops a reality check that'll make you think twice about AI hype. "Outside the Bay Area, GPT-4 changed *nothing*" - and he's not wrong. While we're all talking about AI revolu...
Cline: The Collaborative AI Coder
Saoud Rizwan and Pash from Cline joined us to talk about why fast apply models got bitter lesson'd, how they pioneered the plan + act paradigm for coding, and why non-technical people use IDEs to d...