YOUTUBE
Tech Videos
Sources AI Engineer All-In Podcast Andrej Karpathy Anthropic Apple Apple Developer AWS Events ByteByteGo Computerphile Cursor Dwarkesh Patel EO Fireship GitHub Google Cloud …
Sources
- AI Engineer
- All-In Podcast
- Andrej Karpathy
- Anthropic
- Apple
- Apple Developer
- AWS Events
- ByteByteGo
- Computerphile
- Cursor
- Dwarkesh Patel
- EO
- Fireship
- GitHub
- Google Cloud Tech
- Google DeepMind
- Google for Developers
- Hung-yi Lee
- Lenny's Podcast
- Lex Clips
- Lex Fridman
- Life at Google
- Marques Brownlee
- Microsoft
- No Priors: AI, Machine Learning, Tech, & Startups
- Numberphile
- NVIDIA
- OpenAI
- Perplexity
- Quanta Magazine
- Slack
- The Pragmatic Engineer
- Visual Studio Code
Here is today’s technical daily video digest, tailored specifically for senior engineers to bypass the marketing noise and focus on real engineering and infrastructure signals.
Watch First
Why Humans Can Still Beat AI at Coding - Ryan Greenblatt is the single video most worth your time today. It bypasses the superficial coding demos to explain why current models plateau quickly after initial codebase scans, failing to match a human’s deep, long-term conceptual understanding of massive enterprise repositories.
Highlights by Theme
Developer Tools & Platforms
For local environments, the Visual Studio Code channel’s demo 📌 Show, Hide, & Pin Any Model #vscode #developertools shows how to declutter the IDE model list by pinning or hiding models from GitHub Copilot, local servers, or Open Router configs. To improve developer workflows, How GitHub’s tiny wins team fixes developer paper cuts on the GitHub channel showcases how small details, like pinning contribution guidelines directly to the sidebar, reduce maintainer friction. While Building apps on Google Cloud on Google Cloud Tech is a highly high-level, no-code pitch, Omarchy Quattro Release ⟡ Cursor Tries to Replace Github ⟡ Grokbot Doesn’t Suck ⌁ Syntax Weekly ⌁ on the Syntax channel tracks Cursor’s ambitious attempt to replace GitHub entirely.
AI & Machine Learning
Leading the technical substance, Anthropic’s Co-founder on the Real Bottleneck to Smarter AI on Lenny’s Podcast argues that model intelligence is still primarily governed by compute scaling laws, meaning physical data centers, chips, and power are the main bottlenecks. However, a combined 10x efficiency gain in algorithms and data suggests we could see a 1000x improvement in cost-adjusted model intelligence over the next three years. Operationally, This company has more AI agents than employees on Microsoft details market-research firm Kantar’s rollout of 15,000 active agents—ranging from simple no-code tools to complex orchestrated pro-code systems like an automated HR People Agent. On policy, David Sacks Predicts the Regulatory Capture Playbook to Ban Open Source AI, Step by Step on All-In Podcast warns of a coordinated effort by closed-source players to push safety compliance standards that open-source models run on local hardware cannot technically meet.
Hardware & Infrastructure
On the infrastructure side, The Rising Price of the AI Race | Bloomberg Tech 8/24/2026 and its companion segment Nvidia Customers Brace for Higher AI Costs on Bloomberg Tech report that memory price hikes are forcing OEMs to pass on a 15% price increase for Blackwell and Rubin systems, which require eight HBM stacks at 12 layers of DRAM per GPU. To fund these spiraling costs, companies are undertaking unprecedented capital raises, with SoftBank launching a record ¥1 trillion retail bond to fund OpenAI commitments and Alibaba securing $10.2 billion to back its cloud division. Yet local communities are resisting the physical buildout; Why Data Center Pushback Is a Unique Form of NIMBYism and Odd Lots: Why Many Communities Are Skeptical of the AI Boom on Bloomberg Tech highlight a unique national pushback driven by deep skepticism about whether society actually needs AI, coupled with concerns over local resources like water and electricity.
Everything Else
In corporate and consumer tech shifts, The Truth About the Bezelless Concept Phone on Marques Brownlee highlights Techno’s zero-bezel phone as an impractical ‘form over function’ concept, while Apple Cuts Jobs as It Reprioritizes Around AI on Bloomberg Tech breaks down layoffs on Apple’s Vision Pro teams as engineering roles shift to core Siri AI infrastructure. On national policy and markets, Inside Trump’s Science Agenda: Anti-Science Claims, Fauci’s Damage, DEI & China w/ Michael Kratsios on All-In Podcast details OSTP proposals to run ‘meta-science’ funding experiments like golden tickets to encourage high-risk basic research, while Crypto’s 24/7 Markets Take Aim at Wall Street on Bloomberg Tech tracks Hyperliquid’s $500 billion perpetual swap market and lobbying efforts to trade derivatives onshore.
🎧 We could compile the deep-dive strategy and hardware discussions from these videos into a high-quality audio briefing that you can listen to on the go.