AI & Agents

• 2026-03-12 AI & Agents Best Ways for LLMs to Use Browsers — March 2026

  Opinionated snapshot of how LLMs interact with browsers: MCP servers, automation frameworks, vision-based approaches, and what actually works in production.

• 2026-03-07 AI & Agents Character Design for LLMs: What Screenwriters, Psychologists, and Roleplayers Know That Engineers Don't — March 2026

  Cross-domain research on character and person representation for LLMs, synthesized from screenwriting, psychology, roleplay communities, and production use.

• 2026-03-07 AI & Agents State of the Art: Agent Harnesses — March 2026

  Opinionated snapshot of agent harnesses, orchestrators, and multi-agent patterns as of March 2026.

• 2026-01-31 AI & Agents Photoshop Levels for Semantic Data: Interactive PCA Filtering

  Adjusting embeddings like Photoshop levels for semantic filtering.

• 2026-01-27 AI & Agents Building Coordinated Intelligence: Daily Extraction with Proportional Allocation

  Proportional budget allocation for knowledge extraction systems.

• 2026-01-27 AI & Agents Self-Healing Production - Autonomous Exception Resolution

  Production systems that auto-detect, diagnose, fix, and deploy.

• 2026-01-26 AI & Agents Tree vs Claw: Agent Memory Paradigms

  DB-backed hierarchy vs file-based workspace memory systems.

• 2026-01-23 AI & Agents Agent Shepherding Playbook

  Managing 4-8 autonomous Claude Code agents in parallel using itermctl.

• 2026-01-23 AI & Agents Autonomous Multi-Agent Coordination - Jan 22 2026

  Coordinating 7 agents across music, video, photos, and voice projects.

• 2026-01-23 AI & Agents Multi-Agent Coordination Session - Jan 23 2026

  2-hour session coordinating 7 agents across parallel projects.

• 2026-01-21 AI & Agents Product Backlog Speedrun for Justin

  36 product ideas to 34 tickets in 13 minutes with Claude Code.