Posts on running a business with an AI workforce, plus 14 one-glance explainers — how the workforce is organised, where your data lives, and what the workers can actually do.
askTheodor 0.9.284: give any chat a folder on your Mac and attach files to any message, turn a recorded task into a worker of its own, and put a support assistant you own on your website with one script tag.
Read the post →
askTheodor 0.9.284 adds Automate: record a task the way you always do it, read back the steps it understood, add your rules — and a worker of its own takes it from there, asking before every click.
Read the post →
askTheodor 0.9.282 opens as one calm chat, sorts every model into three cost tiers, tells you when you're on the wrong one — and answers on your Mac's GPU with no account and no network.
Read the post →
A plain tour of askTheodor: a chief of staff named Theodor, workers with memory, plans that run while you're away, and an approval gate between your team and the outside world — all on your own machine.
Read the post →
Almost every company has already tried this once. The demo went fine, everyone was impressed, and then nothing happened. Here's why pilots stall, and the four unglamorous things that make one stick.
Read the post →
An AI worker isn't a chat window with a costume. Here's the six-part definition we build against — and the five things an AI workforce is not.
Read the post →
askTheodor's latest update ships a full marketing team — SEO, content, paid, email, CRO, brand, analytics and more — with 104 marketing skills they already know how to run.
Read the post →One-glance explainers. Click any card to open it as a post — full size, with the story behind it.
Read →
The one-page tour: brief Theodor, he plans and delegates, workers do the work, you approve what leaves — plus what's in the box and where it runs.
Read →
From the 07:30 intake brief to the 18:30 owner report — a full day of work flowing through your local AI team.
Read →
What's inside every askTheodor employee — a soul, a profession, learned skills, a brain you choose, and a tool allowlist.
Read →
A town of workers, run by AI — how Theodor and the specialists are organised into a workforce that delegates like a real one.
Read →
Agents prepare, humans approve: the risk gate, the approvals queue, and the audit log that keep you in command.
Read →
Six questions that decide who owns your office — your data, day one, offline, the model, API keys, oversight.
Read →
82 professions, 252 skills, 116 task templates, 80 tools, 49 providers, 100% local — the product at a glance, verified against the codebase.
Read →
49 LLM providers, one click each — local engines, direct labs, gateways, inference clouds — and your machine at the center.
Read →
The full Claude lineup — tuned, priced and cached correctly, with the right model routed to each job.
Read →
80 built-in tools organised by purpose — files, web, a real browser, ops, documents — governed by least privilege.
Read →
Six doors in and out of your office — MCP presets, connectors, agent presets, A2A — all yours to open or keep shut.
Read →
The business-in-a-box ladder: pick a plan, hire the team, open the playbook, do the real work — up to the first paying customer.
Read →
How an AI chief of staff drives a high-stakes yachting operation: plan, coordinate, execute, monitor, improve.
Read →
The whole story on one sheet — Theodor, the rocket, CEO mode, the city of workers, the trust shield, and friends.
Ready to put a workforce to work?