Skip to content
Shortlist

A hiring studio for hard senior searches

Hiring, with the bias taken out before anyone reads a word.

Shortlist runs hard senior searches. Every search starts blind: paid work samples are scored before names, photos, schools, or dates enter the room. Identity arrives at the interview, after the evidence.

  • 34

    searches closed, last twelve months

  • 91%

    of placements still in seat at month twelve

  • 38%

    of finalists came without a referral

Most hiring decisions are made in the first eight seconds. We take those seconds away.

A reviewer forms a view of a CV before reaching the second line. That view is mostly the name, the photo, and the crest of a university. None of it predicts the work. Our process is built so the first judgment anyone makes is about output.

  • Work before identity

    Every search opens with a short, paid work sample built with you and scored against a rubric we freeze before the first submission arrives. Your panel reads the work of anonymous candidates first. Nobody argues about a person; they argue about the evidence.

  • Structured, or it does not count

    Same questions, same order, same scorecard for every candidate in a search. If an interviewer wants to add a question mid-loop, it goes into the next search, not this one. Consistency is what makes comparison honest.

  • No back-channel notes

    We do not send 'culture fit' notes, 'gut feel' flags, or 'not sure, something felt off' messages. If a concern cannot be written on a scorecard line and defended out loud, it does not travel. That rule has no exceptions, including for founders.

  • Candidates are paid for their time

    Work samples are compensated at a fair day rate whether or not the candidate advances. People doing real work for your decision deserve real money, and it keeps our samples short and our pipelines honest.

The method

The blind first read

This is a sample dossier, exactly as a review panel receives it. The switches control what the panel is allowed to see. When a source of bias is withheld, the signal it carried stays: the degree without the crest, the tenure without the birthday math.

By default everything is withheld, because that is how a Shortlist search begins. Turn a switch off to put that detail back and watch what it adds: nothing the scorecard needs.

What the panel may see

Scripts are off in your browser, so the switches are not shown. What you see above is the dossier in its blind state, the way every Shortlist panel first reads it: name, photo, school, dates, and gaps withheld; the work left in.

Sample dossier. The person, employers, and dates are invented for this demonstration.

No photo

Marta JasikCandidate 07

Staff engineer, payments and billing

Gdansk, works remote

Education

MSc Software Engineering
Gdansk University of TechnologyInstitution withheldClass of 2012Year withheld

Experience

Principal engineer, billing infrastructure

Feb 2024 to now2 yrs 6 mos

Runs invoicing for a nine-country subscription business.

Lead engineer, freight pricing platform

Mar 2019 to Nov 20223 yrs 9 mos

Took a quoting engine from 40 to 2,600 requests per second; team of six.

Senior engineer, card payments

Jun 2014 to Feb 20194 yrs 9 mos

Owned settlement and reconciliation for a two-sided marketplace.

47 / 50

Work sample score

Rubric: correctness, failure handling, written reasoning. Scored blind by three reviewers.

Identity is not hidden forever. Once work samples are scored, finalists meet the panel in person, with the same structured loop for everyone. The point is order: evidence first, identity second. Most processes run it the other way around and call the result instinct.

Searches we can talk about

Employers stay anonymous; that is part of the deal. The numbers are the part we are allowed to keep.

A 60-person logistics software company, series B

Four staff engineers, distributed systems

Nine weeks, first brief to fourth signed offer

The panel scored forty work samples before seeing a single name. The argument in the debrief was about failure handling, not about anybody's employer logo.

  • Median time from application to decision: 19 days
  • All four still in seat at month twenty-four
  • Two of the four hires held no university degree

A national insurer, 400-person product organization

Head of design and two design leads

Eleven weeks, run alongside a panel rebuild

Their previous search had died twice in nine months. The difference was not better candidates; it was a panel that could finally agree on what good meant.

  • Portfolio reviews rebuilt as structured walkthroughs with a frozen rubric
  • Offer acceptance three of three, all at or below the approved band
  • Interviewer scorecard completion went from 61% to 100%

A nine-person climate analytics startup, seed stage

First two data hires, one senior and one mid-level

Six weeks, founder-led panel of two

At that size every hire is a fifth of the company. The founders wanted certainty; what we could give them instead was evidence, which turned out to be better.

  • Work sample: a half-day modelling task on the company's own open data
  • Both hires shipped to production inside their first month
  • Total candidate compensation paid for samples: 4,100 euros

Roles we fill

Senior individual contributors and the people who lead them. If a search is outside this table we will say so on the first call and point you somewhere better.

AreaLevelsTypical searchNotes
EngineeringSenior, staff, principal6 to 9 weeksBackend, infrastructure, payments, data platforms
Engineering managementEM to VP8 to 11 weeksFirst managers a specialty; we test coaching, not charisma
ProductSenior PM to CPO7 to 10 weeksWork sample is a real discovery brief, never a take-home strategy deck
DesignSenior IC to head of design7 to 10 weeksStructured portfolio walkthroughs with a frozen rubric
Data and MLSenior to principal6 to 9 weeksSamples run on your data under NDA, or on public sets
Operations and financeLead to director5 to 8 weeksFor companies past 40 people, first ops leaders included

Writing

The 'culture fit' note is a bias receipt

When a panel cannot name the problem, it names the person. What those four words cost you, in candidates and in court.

June 2026 · 6 min

Structured interviews are boring. That is the point.

The most predictive interview format ever measured is also the least fun to run. Choose accordingly.

April 2026 · 5 min

What a CV cannot tell you, and the little it can

Sixty years of selection research in one honest page: tenure math, credential proxies, and the two lines actually worth reading.

February 2026 · 8 min

Short arguments from the practice. New pieces roughly monthly.

Brief us on the role you keep not filling

Tell us the role, the level, and what went wrong the last time you tried. We reply within two working days with an honest read: either a scoping call within the week, or the name of someone better placed to help.

brief@shortlist.studio

What happens next

  1. A written read on your search within two working days
  2. A 30-minute scoping call; you meet the people who will run it
  3. A fixed fee in writing before anything starts. No percentage of salary, ever