Collection

Library

Browse your videos and channels

Video Assets

919

Tracked Channels

30

Videos 919 Channels 30

AI Engineer

2026 AI Engineer Vibe Reel

1 min read

## At-a-Glance - Verdict: Skip - Core thesis: No substantive thesis is present; the transcript consists only of repeated “Outro Music.” - Why it matters: The source contains no usable information on AI engineering, syste…

Unread

AI Engineer

From Transcription to Live Music: Gemini's Audio Stack — Thor Schaeff, Google DeepMind

13 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Google DeepMind has built a unified audio stack on Gemini 2 that integrates deep audio understanding (transcription + nuance), voice direction (30 base voices modifiab…

Unread

AI Engineer

RAG is dead, right?? — Kuba Rogut, Turbopuffer

10 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Traditional one-shot RAG (vector search → LLM) is being replaced by iterative, multi-tool agentic retrieval that combines semantic search, full-text search, and file-s…

Unread

More or Less Podcast

Would you trade an employee for an agent?

2 min read

## At-a-Glance - Verdict: Skip - Core thesis: Current AI agents lack continuous learning and improvement compared to human employees, making them insufficient replacements despite their utility - Why it matters: Frames t…

Unread

Stripe

Bringing world-class payments to UK small businesses

1 min read

## At-a-Glance - Verdict: Skip - Core thesis: Lloyds Banking partnered with Stripe to launch Lloyds Accept, a mobile card payment tool for UK SMBs, emphasizing brand trust and cash flow improvements for service businesse…

Unread

Every

Anthropic’s Fable 5: A Warp Drive for Coding

17 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Anthropic's Claude 5 (marketed as Fable) is a 'warp drive' model that compresses months/years of autonomous coding work into hours/days, scoring 91/100 on senior engin…

Unread

AI Engineer

Why Eval++ Is the Next Great Compute Primitive — Sunil Pai & Matt Carrie, Cloudflare

21 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Dynamic Workers (eval++) and Durable Objects provide a fundamentally better compute primitive for AI agents by enabling stateful serverless execution with secure code …

Unread

AI Engineer

Why More Context Makes Your Agent Dumber and What to Do About It — Nupur Sharma, Qodo

18 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: More context in agentic systems causes LLMs to lose focus on the middle of large inputs (U-curve attention), requiring strategic context optimization and multi-agent a…

Unread

AI Engineer

Road to 5 Million Tokens: Breaking Barriers in Long Context Training — Max Ryabinin, Together AI

9 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Together AI achieved 5 million token context length training on 8x H100 GPUs by stacking existing memory optimization techniques and introducing 'UPipe' (chunked atten…

Unread

20VC with Harry Stebbings

Nebius Co-Founder on AI Infrastructure Bubbles | How Price Elastic is Demand for Compute

48 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: AI infrastructure demand is in early innings; vertical integration from bare metal to managed inference and customer diversification insulate Nebius from commoditizati…

Unread

AI Engineer

Building Interactive UIs in VS Code with MCP Apps — Marlene Mhangami & Liam Hampton, GitHub

12 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: MCP apps extend Anthropic's Model Context Protocol to render rich, interactive HTML UIs directly in chat windows (like VS Code's GitHub Copilot), eliminating text-only…

Unread

AI Engineer

From MCP to Scale: Pipelines That Build Themselves — Rafael Levi, Bright Data

17 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: LLM agents can now build, run, and self-heal web scrapers at scale using Bright Data's MCP (Model Context Protocol), saving ~62%+ tokens vs. parsing raw HTML and elimi…

Unread

AI Engineer

Under 5 minutes to a deployed LLM endpoint — Audry Hsu, RunPod

8 min read

## At-a-Glance - Verdict: Skim - Core thesis: RunPod provides rapid GPU deployment for LLM endpoints via a hub of pre-configured repos, serverless auto-scaling, and a pay-per-second model targeting developers who want in…

Unread

AI Engineer

Text Diffusion — Brendon Dillon, Google DeepMind

22 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Text diffusion models achieve 10x lower latency than autoregressive models by iteratively refining entire token sequences in parallel instead of generating one token a…

Unread

AI Engineer

Building safe Payment Infrastructure for the autonomous economy — Steve Kaliski, Stripe

13 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Agents spending money require deterministic payment/checkout flows with scoped credentials and structured negotiation, not browser automation—Stripe built three primit…

Unread

AI Engineer

Evals Are Broken, Use Them Anyway — Ara Khan, Cline

15 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Traditional benchmark evals are misleading at best and fraudulent at worst, but when combined with real-world agent testing, continuous hill climbing, and vibes valida…

Unread

20VC with Harry Stebbings

Why Token Maxing is Failing Enterprise Startups | Legora CTO

45 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: AI-powered development has shifted the engineering bottleneck from code-writing to product definition and code review; competitive advantage now depends on org archite…

Unread

AI Engineer

Beyond Transcription: Building Voice AI That Understands Conversations — Hervé Bredin, pyannoteAI

16 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Speaker diarization (who spoke when) is critical infrastructure for multi-speaker transcription, but state-of-the-art still fails badly in noisy/overlapping speech; py…

Unread

AI Engineer

Dark Factory: OpenClaw Ships Faster Than You Can Read the Diff — Vincent Koc, OpenClaw

14 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: OpenClaw ships 800–3,000 commits/day by treating AI agents as factory workers, not magic—engineering discipline, parallel 'swim lanes,' .skills management, and human f…

Unread

AI Engineer

Building Agent Interfaces: Lessons from Chrome DevTools (MCP) for Agents — Michael Hablich, Google

15 min read

## At-a-Glance - Verdict: Watch fully - Core thesis: Building agent interfaces requires treating agents as a separate user segment with distinct cognitive bottlenecks (token burn, error recovery, tool discoverability, tr…

Unread