How YC startups use AI for sales

Sales is where AI stopped writing drafts and started running the motion. The clearest proof is Y Combinator's own portfolio: more than a decade of startups, from the first email-follow-up bots to today's autonomous reps, building the AI-native version of selling.

Below are ten of them, what each one automates, the patterns they share, and how to copy the playbook in your own startup. Company names and batches are public on Y Combinator (see Sources).

The shift: from a bigger team to a faster loop

The old sales model scales with headcount. More pipeline means more SDRs researching accounts, writing emails, chasing follow-ups, and updating the CRM. The AI-native model breaks that link: an agent does the research, the writing, and the chasing, and your people spend their hours on the conversations that actually need a human.

Three shifts make that possible, and you can see all three in the companies below: outreach gets triggered by real buying signals instead of a static list, the agent acts (researches, emails, books, logs) instead of just drafting, and the newest wave moves AI onto the live call itself, coaching the rep in real time rather than scoring the recording afterward.

Ten YC startups building AI for sales

What they have in common

• They start from signals, not lists: the trigger is a job change, a website visit, or an intent spike, so the first touch lands while the buyer is actually in market.
• The agent acts, wired into real tools (email, calendar, CRM). It researches, sends, books, and logs, instead of handing a rep a draft to copy and paste.
• The newest wave moves onto the live call: real-time coaching during the conversation, not a QA score read the next morning when the deal is already cold.
• Most are narrow on purpose: a specific motion (outbound, ABM, inbound) or a specific industry. For sales, a focused agent outsells a general one.

How to copy this in your startup

1. Pick one motion to automate first, usually top-of-funnel: research plus the first cold email. That is the most repetitive, lowest-judgment part of selling and the easiest win. This is the heart of Module 6 (Convert).
2. Trigger outreach on a signal you can actually detect (a website visit, a job change, a product action), so the AI contacts buyers who are in market instead of blasting a static list.
3. Give the agent least access first: read-only on your CRM and inbox, then let it draft and send only after you approve, before you let it act on its own. Treat it like a new SDR, not a trusted admin.
4. Keep a human on the live conversation and the close. Let AI handle research, first touch, and follow-up; let your judgment handle the deal once a person replies.

Building support this way (AI that resolves, with a human in the loop and least-access by default) is exactly Module 6 (Convert) of AI Operating System for Startups.