We test whether they can handle a customer, too
Plenty of engineers can code. Far fewer can sit with a frustrated customer, work out what they actually need and ship it by Friday. We test candidates on both before you meet them. You run the final interviews and make the hire.
Free to interview. 90-day replacement guarantee.
Every candidate you see has passed steps 1 and 2. Deployment strategists do a business case in place of the technical screen. Candidates go through this core vetting once and their scores carry over between searches. For your search, we start by scoping the role with you, then run a short second role-play with a customer modeled on yours. If you sell to hospital IT, the "customer" is a hospital IT lead. We score every shortlisted candidate against the bar we agreed with you. Steps 3 and 4 are there if you want more evidence before your final interviews.
Technical screen
The candidate gets one task: take a messy, badly documented API and get working code running against it. For AI roles we hand them a retrieval pipeline that returns wrong answers instead, and they have to find out why and prove the fix with a small eval. We pay most attention to how they read code they didn't write and what they do when the docs turn out to be wrong.
What we score: working code · debugging · reading other people's systems
Customer role-play
An FDE plays the customer, usually an operations lead with a vague problem, a skeptical VP and a deadline. The candidate has to work out what the real problem is, set expectations they can meet and turn down the wrong request while keeping the customer on side. A resume tells you very little about whether someone can do this, so we watch them do it.
What we score: discovery questions · clarity · pushback · composure
Live build
The candidate turns the problem they scoped in the role-play into a working demo and presents it back to the "customer." It works like a compressed first deployment, and it shows you what they cut when time runs short.
What we score: shipping under pressure · judgment · the demo itself
Customer references
We call their managers and, where we can, the customers they deployed for. The question we always ask is whether they would want this person back on their account. If a reference makes us nervous, we tell you before you make an offer.
What we score: trust · follow-through · how they handled a bad week
What you get with each candidate
Every shortlisted candidate comes with a scorecard written against your bar, including how they did in the role-play built around your customer. We write down what worried us as well as what impressed us.
- Scores for the technical screen and the customer role-play
- Our concerns, written plainly
- What we'd probe in your final interviews
- Availability and notice period
What we don't do
We don't set whiteboard puzzles
Inverting a binary tree tells you nothing about how someone will handle a deployment. Everything we test is work an FDE does in their first week.
We don't filter on pedigree
Some of the best FDEs came from support, consulting or solutions roles. We screen on what people can do, and we don't care which logos are on their resume.
We don't forward resumes
If we haven't tested someone ourselves, we don't send them to you. We'd rather send you three people we'd stand behind than thirty you have to sort through.
See who clears your bar in under 5 days
Interviewing is free, and every hire comes with a 90-day replacement guarantee.