Founding Technical Recruiter

Every company says it listens to its customers. Few actually talk to them, because real interviews take weeks to recruit, schedule and analyze, and the answer usually arrives after the decision has been made.

Dialogue fixes that. Our AI interviewer holds live, in-depth voice conversations with real people, recruited from a vetted panel of more than ten million, in 87 languages. It follows up the way a good researcher would, and the findings land in hours instead of weeks, with every insight linked to the quote and the video behind it. Teams at Meta, TikTok, Wayfair, Tonal and Kalshi use it to test concepts, understand why customers leave, track their brand and decide what to build next. Thousands of interviews run through it every month.

We were founded by product leaders from Apple, Reddit, Snapchat, Twitter and Nextdoor, our research team comes from Airbnb, Slack and Google, and we are backed by Lightspeed. We are one of the fastest-growing startups in Los Angeles.

About the job

This is the most important hire we are making ahead of hypergrowth. We have the customers and the product. The constraint on the next year is how many great engineers we can get into the building, and you are the person who removes it.

You are our first in-house recruiter. You report to our CTO and sit next to him. Our VP of Growth and BizOps, who owns people, runs the weekly funnel review with you and makes every offer. The mandate is ten engineers in Los Angeles in the next six months, then the next twenty people across research, go-to-market, design and finance. Every engineer works in person in West Los Angeles, five days a week, and most of the ones we want do not live here yet. Part of this job is bringing them here.

We are not necessarily hiring a recruiter with a network. We are hiring a recruiter who runs a search the way a growth team runs a funnel: a pool measured in thousands, scored before the CTO sees it, outreach tested daily, every stage timed, and a Monday dashboard that says where the candidates are and what each of our constraints costs us.

In one week you might calibrate the rubric with the CTO against the last ten people he said yes and no to, source and enrich three hundred engineers in San Francisco and at the top computer science programs, test two outreach variants and kill the loser by Wednesday, run eight phone screens, move three to the programming screen, chase every open next step in the ATS, and prep the founders for two closing calls.

What you will own in your first 90 days

• The rubric. Our CTO knows a great engineer when he sees one and will tell you the bar is hard to write down. You write it down with him, test it against his past decisions, and make it the score every profile carries before he sees it.
• The pool. A sourced, enriched and scored base of engineers who are in Los Angeles or would move here, drawn from San Francisco and NYC, the top computer science programs and the places strong early-career engineers actually are: hackathon results, open-source contributors, YC's hiring lists, the group chats. Thousands, not hundreds, refreshed weekly.
• Outreach. Personalized at scale from a real mailbox, measured daily: sent, opened, replied, booked. Variants tested, losers killed, winners written down.
• The phone screen. You run it. Can they name exactly which model and effort level they code with. What do they read to keep up. Can they explain their work without rambling. You pass the ones who clear it to the programming screen and tell the CTO why.
• The process. Every candidate with a next step and a date, interviews scheduled within 48 hours of a yes, candidates told where they stand. Speed is a feature. We lose people without it.
• The close. The founders sell and sign. You set them up: the candidate's situation, what they care about, the relocation package and the comp conversation ready before the call.
• The dashboard. One page the founders read every Monday: pool size, touches, replies, screens, onsites, offers, accepts, and days at each stage.

The bar is quantitative

We will ask for four things, and we mean all four.

• Your funnel. From a search you ran: sourced, contacted, reply rate, screens, onsites, offers, accepts, time to fill. If you cannot produce it, you did not own it.
• Your tooling. Something you built, not a default you used. A scripted enrichment, a scoring sheet, an outreach experiment with a result. Gem and LinkedIn Recruiter are table stakes, not the bar.
• Your model. Which AI model and effort level you use for what, and why. "I use Claude to draft emails" is the wrong answer here, the same as it is for the engineers you will recruit.
• Your work sample. Every finalist does the same one, whatever their background.

Where it goes

Engineering first, then everyone. Once the engine runs for engineers you own the hiring plan, and the person who does this well ends up running talent at a company several times this size. We are a meritocracy. Scope follows judgment, not tenure, and we do not see a ceiling on this role.

You might be a fit if

• You have 3 to 7 years in in-house technical recruiting or sourcing, at least some of it at a company that measured the funnel and held you to it.
• You have closed engineers into five-day in-office roles, and you can say which arguments worked and which did not.
• You know your numbers cold and you are a little embarrassed by recruiters who do not.
• You have built your own tooling and can show it: Python or SQL, a Clay or Apollo pipeline, a scraper, a sheet that scores.
• You write short, specific, human cold messages, and you can tell within a day whether a change moved the reply rate.
• Your network skews to the engineers we are hiring: one to five years in, from strong programs, shipping with agents. Not only the staff-level engineers you placed five years ago.
• You can hold a technical conversation with an engineer without pretending.
• You like being measured. Your funnel is public inside the company every week.
• You close a hundred small loops on time. Recruiting is mostly that.

AI fluency

This is the job, not a nice-to-have. Our CTO passes on candidates who cannot name the model and effort level they use. The recruiting engine you build runs the same way.

• You use Claude or its peers agentically: scraping and enriching, scoring profiles against a rubric, drafting and personalizing outreach, turning a screen into a scorecard, keeping the ATS honest.
• You check what an agent hands you. A confident wrong score or an invented employer costs us a candidate.
• When you do something twice, you turn it into a skill, a script or an automation, and you write it down so someone else can run it.
• You have built something with these tools that you can show us. Bring it to the interview.

How we work

In person, five days a week, in West Los Angeles. Everything is written down. We will tell you what good looks like once, in writing, and expect you to run it from there. You will have the ATS, the sourcing stack, a cold outbound mailbox and a data budget from week one.

Pay, benefits and logistics

Los Angeles, in person. Reports to the CTO. Start as soon as possible. If you are not in Los Angeles, we will help you get here.

• Base salary $165,000-180,000, plus equity
• Performance bonus of up to $50,000 to $75,000 annually
• Full health, dental and vision coverage
• Flexible PTO
• Office meals and premium snacks
• Productivity stipend
• Team events and offsites
• Commuter benefits
• Relocation support

How to apply

Have the funnel from your last engineering search ready. It is the first thing we ask for, and rough numbers are fine. No cover letter. We reply to every application.

Finalists do a two-day paid work sample at home: build a sourced and scored list of fifty engineers who meet a spec we give you, write five outreach notes, and tell us what it would take to get to two thousand.