THE BLOG
Field notes on running a business with AI.
What's working, what isn't, and what we're learning building TigerTiger in the open.
October 19, 2026
The Territory #12: Claude Code idle chats, OpenAI's agent incidents and Gemini 4
Claude Code now trims idle chats, OpenAI lists 19 times its agents went off script, and Google releases Gemini 4 Argon.
Read more →
October 12, 2026
The Territory #11: Claude's dashboards, Google add-on and Claude Code advisor
Claude Code's advisor setting now works with the small Haiku 5.5 model, and Claude can edit Google Docs, Sheets and Slides.
Read more →
October 6, 2026
The Territory #10: Two Claude Code updates
A second assistant that points out what you might miss, and a check for instructions written for older models.
Read more →
September 24, 2026
The Territory #9: A new AI for the small calls in your workflows, plus 2 Claude Code updates
We wired a new decision-making AI into our workflows. Here's what it caught, and where it sounded sure and was wrong.
Read more →
September 14, 2026
The Territory #8: A security report, and three smaller changes
Anthropic published what criminals are doing with Claude, Claude Code costs can now match your admin's numbers, ChatGPT lets you sketch instead of describing, and Google shipped a third fast Gemini model since July.
Read more →
August 26, 2026
Four Hours From Zero: An AI Agent Workshop at Spiderhook
Company heads and function leads at Spiderhook built a working AI agent workflow in a single morning. Here's what the session covered and why leadership was in the room.
Read more →
July 28, 2026
Your Old AI Instructions Are Making the New Models Worse
Anthropic just deleted over 80% of the instructions it gives its own AI coding tool, and performance got better. Here's what that means for the instructions your business is running.
Read more →
Put it to work.
TigerTiger runs inside your own Claude Desktop. Join the waitlist and we'll reach out when a spot opens.
Get in touch