Every founder, ops lead, and support manager loses 2+ hours a day to inbox triage — classifying, drafting, routing. Our inbox agent watches Gmail or Outlook, uses Claude Haiku to classify every email (support, sales, invoice, urgent, noise), Claude Sonnet to draft a reply in your voice, and WhatsApp to show you the draft before it sends. CRM auto-updates. $0.10 per email resolved. Noise is free.
Assumes linear scaling and no ramp-up. Real deployments hit steady-state around week 3–4.
Only if you explicitly enable autonomous mode for a specific category (e.g., 'auto-reply to known vendor invoice acknowledgements'). Default is human-in-the-loop for every send. Every send is hash-chained for audit.
Set excluded senders (your boss, your spouse, your investors) or excluded threads (the offer letter negotiation). The agent skips those and leaves them in your normal inbox unchanged.
On day 1 we point Claude at your last 60 days of sent mail. It builds a voice fingerprint — sentence length, common phrases, salutation style, sign-off patterns. You can correct it in-line during the first week.
No. We use the Anthropic API with the no-training contractual default. Email content is sent for inference only. See /compliance.
Our inbox agent handles moderate volume (low hundreds per day per tenant). For Sephora-scale tier-1 support (10K+ tickets/day, deep Zendesk integration, brand-voice consistency), Sierra or Decagon are stronger picks — see /vs/sierra and /vs/decagon.
A $0 outcome audit takes 3 minutes. We'll model your monthly savings against your real volume — and show you what week-one looks like.