A build log for one real Tamaton custom agent that triages the inbox, checks the calendar and docs, drafts replies, and files attachments — plus the three failure modes we fixed.
Coding demos are graded by a test suite. Email is graded by your boss, your customer, and your calendar. Here's why the inbox is the hardest honest test of an AI agent.
AI agents aren't human users or dumb service accounts — they're a third identity class. Here's how to issue agent credentials and scope permissions before your inbox pays for it.
A concrete walkthrough of building a Tamaton agent that triages inbox, drafts replies with real context, and books follow-ups — with a human holding the approval button.
Inside the eval harness we run on inbox triage, doc drafting, and calendar scheduling — the golden sets, the scoring rubrics, and the failure modes we refuse to ship.
A concrete walkthrough of Tamaton's inbox triage agent: how 84 unread messages become a ranked queue with searches run, drafts written, and calendar holds already placed.
Shared API keys give bots the run of your whole workspace. Tamaton issues each agent its own identity with per-layer, per-resource permissions instead.
Task-completing agents can't be graded like chatbots. Here's an eval framework built around side effects, reversibility, and the cost of being wrong.
The bottleneck for multi-agent setups isn't a smarter model — it's shared state, clean handoffs, and knowing who owns the calendar. Coordination is the real work.
Move past one-off prompts. Build a persistent Tamaton agent that watches your inbox, files, and calendar, then acts on a schedule with human checkpoints.
A scheduling bot books slots. A real AI calendar agent remembers, decides, and negotiates. Here's how to design the second one.
A product deep-dive on the permission, preview, and rollback design that makes an AI email agent safe to actually act — not just suggest.
Get started
Claim your address before someone else does — free to start, with an AI-native inbox built in.