Notes from building Distill — what's shipping, what's breaking, what we're learning about skills, agents, and the future of working with AI.
The AI productivity landscape is splitting into two camps: general assistants and workflow-specific tools. Here's what actually moves the needle.
Screen-aware AI watches your actual workflow, extracts repeatable patterns, and builds skills it can replay. Here's the current state.
A practical breakdown of AI workflow tools in 2026: screen-aware recorders, context-learning assistants, and what actually delivers.
AI workflow tools now observe your screen, extract the repeatable steps, and turn them into reusable automations. Here's how it works.
Designers repeat dozens of mechanical steps daily. AI can now observe those patterns and automate the tedious parts. Here's what's working.
Most AI tools start from zero every session. These ones build persistent memory of how you work and compound over time.
AI memory comes in three levels. Only one of them actually remembers your workflow. Here's the state of play in 2026.
Turn your recorded work sessions into repeatable, parameterized workflows. A step-by-step guide for developers and knowledge workers.
You automated your tests, deploys, and linting. You still spend half your day on repetitive work. Here's why — and what to do about it.
Skills extraction turns raw workflow recordings into structured, parameterized skills. Here's how the technology works and where it's headed.
First post on the Distill blog. Notes on what we're building and why.