AI videos, nothing else.

Claude Skills: Storage and Context Limits

The alexkerss channel demonstrates how skill files, references and context limits can affect Claude workflows.

Parv Mahajan on Measuring AI Research Acceleration

Parv Mahajan discusses uncertain evidence about AI’s contribution to research acceleration.

Claude Code Updates for Background Work and Compaction

AICodeKing reviews recent Claude Code features and their implications for coding workflows.

Choosing Execution Boundaries for AI Agents - Mislav Ivanda

Mislav Ivanda argues that execution boundaries deserve attention before choosing an AI agent framework.

Pratyusha Duvvuri on the Cost of AI Tokens

Pratyusha Duvvuri explains why AI token costs are difficult to track across providers, teams and finance structures.

Jason Mancuso on Scaling Agent Reinforcement Learning

Jason Mancuso explains how Modal approaches reinforcement learning for agents across distributed GPU infrastructure.

Building a Self-Improving Software Factory - Zach Lloyd

Zach Lloyd outlines how teams can measure and improve agent-assisted software development as an infrastructure system.

Richard Feldman on Sharing Agent Work and Context

Richard Feldman explores how teams can retain context when collaborating on agent-driven code changes.

Jemima Abu on Accessibility for AI Coding Agents

Jemima Abu explains how to guide coding agents toward more accessible products and why prompts alone are not enough.

Ashwin Paranjape on Testing AI Earnings Forecasts

Ashwin Paranjape describes a controlled evaluation of AI agents predicting corporate earnings.

The Context Coding Agents Cannot See - Tom Moor

Tom Moor explains how carrying product context through development can help coding agents make more coherent changes.

Why Coding Agents Need a Well-Tended Codebase - Theo Browne

Theo Browne compares maintaining a development environment for coding agents to tending a garden.

Matthew Jewkes on Routing Exceptions in AI Automation

Matthew Jewkes explains how routing routine work and exceptions can focus expert attention in AI automation.

Keiji Kanazawa on Reviewing Agent-Written Code

Keiji Kanazawa recounts an agent-driven refactor that failed and explains how he rebuilt the project with evaluation loops.

Testing an AI Forecast Against a Market Sell-Off

All About AI tests whether a model could have anticipated a market decline using information from before it occurred.

Angela McNeal on Controlled Autonomy in Enterprise AI

Angela McNeal explains why enterprise AI workflows need clear decision logic, oversight and records.

Riley Brown Connects Claude Duet to Local Coding Sessions

Riley Brown demonstrates sending requests from Claude Duet to local Claude Code sessions.

How Companies Change Workflows for AI Coding - Martin Harrysson and Prakhar Dixit

Martin Harrysson and Prakhar Dixit discuss organizational changes they see in enterprise AI coding workflows.

Ryan Stevens on Evaluating an AI Bookkeeping Agent

Ryan Stevens explains how Ramp tests a bookkeeping agent using synthetic businesses and end-to-end evaluations.

Natalia Connolly and Kevin Madura on AI Procurement Analysis

Natalia Connolly and Kevin Madura explain how structured data and verification improve AI procurement analysis.