Nine NDAs triaged against your playbook by lunch
A companion sorts contract intake into three piles with the deviations highlighted, chases the six unsigned agreements, and builds the renewal calendar nobody has.
Nine NDAs sorted against your playbook by lunchtime: the ones that match, the ones deviating into a fallback you already approved, and the ones needing a lawyer, with the deviations highlighted. Six unsigned agreements chased. And a renewal calendar that finally exists, with every notice deadline in it.
Most legal ops teams are one person, and the queue is the job. A companion works it in your real accounts. PandaDoc is a native integration, and so are Notion, Google Docs, Google Drive and Airtable, where intake trackers and clause playbooks live. DocuSign, Ironclad and whatever CLM you run get worked in your logged-in session.
What does AI-assisted NDA intake look like?
The counterparty’s paper arrives. A companion reads it against your playbook, the ten positions you actually care about (governing law, term, non-solicit, IP carve-outs), and sorts the queue into three piles: matches, deviates within a fallback you pre-approved, and needs a lawyer. The first pile turns a Thursday turnaround into a same-day one. Nothing is signed, sent or conceded by the companion. It drafts the redline notes and the routing, and the judgment pile reaches a human desk with the deviations already highlighted.
Chase the six unsigned agreements
From PandaDoc’s document list: everything sitting sent-but-unsigned for more than five days gets a drafted nudge naming the document and the waiting party, and the batch waits for your approval. If your signature flow runs on DocuSign, the same chase works through your session. Six nudges reviewed in a minute, every week, without you having to remember to look.
Can AI build and keep the renewal calendar?
This is the one that pays for the setup. A companion walks the contract repository in Drive or your CLM session, extracts term, auto-renewal language and notice deadlines into an Airtable or a sheet, and a scheduled routine surfaces anything with a decision window opening in the next 90 days. The vendor-renewal playbook shows the shape. Scanned paper comes back with the extraction marked for a paralegal to confirm, which turns a week of reading into a day of checking.
What stays with the lawyers
Advice, negotiation, and anything that constitutes practicing law. Sorting nine NDAs against a playbook a lawyer already wrote is the half a companion carries. If you are a lawyer rather than an ops function, the lawyer page goes deeper. Which repositories a companion reads is a per-companion choice, so privileged material can stay out of scope entirely; the security model sets out the rest.
Experience Strawberry for free
DownloadTrusted by fast-growing companies worldwide
Frequently asked questions
Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents