A role that sits open for months usually has the same root cause
The team blames the market, the compensation band, or the recruiter. The real problem is almost always upstream. Nobody wrote down what the role actually was before the search started.
Strong AI teams treat that definition as the first deliverable of a hire, not paperwork to backfill later. It is the difference between a search that reaches the right candidates quickly and one that produces resumes without a clear standard.
Start with outcomes, not a requirements list
A typical job description lists technologies and years of experience. A useful role document answers three questions instead. What will this person ship in their first 90 days? What stack will they touch every day? What does failure look like in this role?
Those answers shape where you source, what you screen for, and how you run the debrief. For example, if the role is about putting models into production on infrastructure you already run, the search should center on deployment experience, not published research.
Name the exact role before you open the search
“AI engineer” covers at least three different jobs. An ML engineer who ships and maintains models is not a data scientist who frames the problem, and neither is the data engineer who keeps the pipelines alive. A brief that does not name which one you need produces a pipeline of people who all look plausible and none of whom fit.
Write the one role you need this quarter.
The hidden cost of skipping the document
When the role is vague, interviews turn into fresh debates about what you are looking for. Debriefs turn into opinion contests. Good candidates drop out while the team argues. The cost is a seat that stays empty and a roadmap that slips.
This is the judgment error worth naming. Leaders treat the document as overhead and the interviews as the real work. It is the reverse. A clear document is what makes the interviews worth running, because it gives interviewers a shared standard and lets you decide quickly when the right person appears.
Where Tesoro AI fits
This document is exactly what we use at the start of an engagement. We run human-led screening against your specific role requirements, confirm bilingual capability and Americas time-zone alignment, and handle compliance, payroll, contracts, and onboarding. Tesoro AI focuses on the top 5% of LATAM AI engineers, with the first pod sourced in 7 days or less and full pods delivered in under 30 days. Costs run 50 to 70 percent lower than US onshore hires.
The sharper your role document, the sharper the shortlist we hand back.
Start here
Before your next AI search, write the one page: 90-day outcomes, the daily stack, and what failure looks like. If you want to turn that document into a shortlist of role-screened LATAM AI engineers, start a Fit Call.