The founding team spent two decades recruiting. A role opens. Hundreds of applications land. A recruiter can only interview a few of them. The first round turns into a queue. Strong candidates wait days for a slot and take another offer. Weak candidates use up the slots that are left. The write-up at the end is a paragraph from memory, and a hiring manager cannot audit that.
More screening software was not the answer. Keyword matching does not tell you whether someone can hold a system design conversation. A resume does not tell you whether the code was theirs. Anyone can sound qualified in writing. So put the candidate in a real round and keep the evidence.
The round came first. It runs 24/7, on the candidate's schedule and in their language, with an AI interviewer named Ethan who asks follow-ups instead of reading a script. Then scoring: you pick the skills for the job and set a weightage on each one, you set up the job description, and answers are scored against both. Then the evidence. The recording, the transcript and the code travel with the score, so a person can check the work.
Everything since is the same idea, pushed outward. Recruiters wanted their own rounds, so human-led interviews land on the same reportcard. Agencies wanted many clients in one place without pooling data, so Clients arrived. Universities wanted a readiness picture before placement season, so the campus side was built. Students wanted to practice against the real thing, so practice uses the same scoring their interviewers will.
