Skip to content
TailoredMinds

Future of hiringGuideUpdated 5 min read

Skills-based hiring for engineering teams: a practical guide

Dropping a degree requirement is the easy part. Here's how to define the skills, gather evidence of them and make decisions on it.

Written by

Lucas Salvaia, Co-Founder & COO

Reviewed by

Tercio Lima, Growth Lead

The short answer

Skills-based hiring means deciding who to hire on evidence of the skills a role needs, rather than on proxies such as degrees, years of experience or previous employers. For engineering teams, that means defining the skills from the actual work, asking every candidate for evidence of them, and assessing them with structured interviews and short work samples. Removing a degree requirement from the job ad is only the first step; most of the change happens in screening and interviews.

Key takeaways

  1. Skills-based hiring replaces proxies (degree, years, brand-name employers) with evidence of the specific skills the role needs.
  2. Research on selection methods finds that years of experience and years of education predict job performance weakly, while structured interviews and work samples do better.
  3. Many employers have dropped degree requirements on paper without changing who they hire. One study estimated fewer than 1 in 700 hires in 2023 resulted from it.
  4. The change has to reach screening, interviews and the final decision, not just the job ad.
  5. Engineering is well suited to skills-based hiring because the work produces evidence: code, designs, systems and incidents.
  6. People still decide. A skills framework makes their judgement more consistent; it doesn't replace it.

What skills-based hiring means

Skills-based hiring is a way of selecting people based on demonstrated evidence of the skills a role requires, rather than on credentials or pedigree used as proxies for those skills. In practice it has three parts:

  1. A clear list of the skills the role needs, derived from the work the person will do.
  2. A way to gather evidence of each skill from every candidate: past work, a work sample, structured questions.
  3. A decision based on that evidence, scored the same way for everyone.

It doesn't mean ignoring experience. It means asking what someone did and can do, rather than counting how long they did it or where they studied.

Why proxies fall short

Degrees and years of experience are weak predictors of how well an engineer will do the job. In Schmidt and Hunter's widely cited 1998 meta-analysis of 85 years of selection research, years of job experience had an estimated validity of .18 for predicting job performance and years of education .10 [3]. A 2022 re-analysis by Sackett and colleagues ranked structured interviews (.42), job knowledge tests (.40) and work sample tests (.33) among the stronger predictors [2].

In engineering, proxies fail in specific ways:

  • Years of experience with a technology say little when the technology is new. Modern LLM application work is only a few years old, so a "5+ years of LLM experience" requirement filters out most of the people who have actually done it.
  • Degrees screen out strong engineers who learned through work, open source or self-study, and say little about whether someone can run a system in production.
  • Brand-name employers tell you someone passed another company's process, not what they built there.
ProxyWhat it actually tells youSkills-based alternative
Computer science degreeCompleted a programme of studyEvidence of the specific fundamentals the role uses, tested in a work sample
"X+ years of experience"Time in a role or with a toolWhat they built, owned and fixed in that time
Well-known previous employerPassed another company's hiring barThe systems they personally built and the decisions they made
Long list of tools on a CVExposure to toolsDepth in the two or three tools your team actually uses
Polished CV wordingWriting skill or AI assistanceA conversation about one system in depth

The gap between intent and practice

Many employers have announced skills-based hiring; far fewer have changed who they hire. A 2024 study by Harvard Business School and the Burning Glass Institute estimated that fewer than 1 in 700 hires in 2023 resulted from employers removing degree requirements. About 45% of the companies that made the change appeared to make it in name only, with no meaningful difference in who they hired [1]. At the firms the study classed as skills-based hiring leaders, workers without degrees had retention rates about 10 percentage points higher than colleagues with degrees [1].

The lesson for engineering teams: editing the job ad is not enough. If the CV screen, the interview questions and the final debrief still reward the old proxies, nothing changes.

Governments are moving the same way. In the United States, the Chance to Compete Act, signed in December 2024, requires federal agencies to use technical assessments, developed with subject-matter experts, for most competitive-service hiring rather than relying on self-reported qualifications [4].

How to run skills-based hiring for an engineering role

Step 1: Define the skills from the work

Start with what the person will do in the first six to twelve months, then derive the skills. For an engineer joining a payments team, "designs idempotent APIs", "debugs production incidents across services" and "writes clear design notes" are skills. "Bachelor's degree" and "5 years of Java" are not.

Keep the list short: four to six must-have skills, each one observable. If you can't describe what evidence of a skill would look like, it isn't specific enough yet. For AI roles, our AI role profiles list the core skills and the signals we look for for each role.

Step 2: Rewrite the job ad around the skills

Describe the work, list the must-have skills and say how you'll assess them. Remove requirements that are proxies rather than needs. Our LLM engineer job description template shows the format for one role.

Step 3: Screen on evidence, not keywords

Read CVs and profiles for evidence of each must-have skill: systems built, problems solved, outcomes owned. A short structured first conversation, with the same three or four questions for everyone, gathers more evidence than a keyword match. Search tools that read skills rather than exact keywords help here, as long as a person reviews who is shortlisted.

Step 4: Assess with work samples and structured interviews

Use a short work sample that mirrors the real job, and a structured interview where every candidate gets the same questions and the same scoring guide. Keep take-homes to two or three hours at most, so you aren't filtering on who has the most free time. Our interview framework for AI engineers works for most engineering roles.

Step 5: Decide on the evidence

Each interviewer scores each skill independently, with written evidence, before the debrief. In the debrief, discuss the evidence skill by skill. If someone argues from a proxy ("they only have two years of experience"), ask what the evidence on the relevant skill shows.

Step 6: Check the results

After a few hiring rounds, look at who got through each stage and why candidates were rejected. If one stage rejects most candidates without a skills-based reason, that's where the old proxies are hiding.

Common pitfalls

  • Changing the ad but not the process. The most common failure, and the one the research above documents.
  • Too many "skills". A list of twenty skills turns back into a keyword list. Keep four to six must-haves.
  • Unstructured interviews. Without shared questions and a scoring guide, interviewers drift back to proxies and gut feel.
  • Long unpaid take-homes. They measure available time as much as skill.
  • Letting a tool decide. AI can help read skills from profiles, but people should review every shortlist and make every decision. Our article on human-in-the-loop hiring covers where to draw the line.

How Tailored Minds helps

Skills-based hiring is how we shortlist. When you brief a role on Tailored Minds, we turn it into a precise profile built around the work and the must-have skills. Our team searches the talent network with AI-assisted search that reads skills rather than keywords, and approved independent recruiters can work the role at the same time. People review every candidate we shortlist against the signals for that role: what they shipped, how they measured quality, what broke and how they fixed it. Talent profiles include an optional intro video, which shows how someone thinks and communicates in a way a CV can't. Hire engineering teams with Tailored Minds, or read our guide to hiring AI engineers.

Frequently asked questions

What is skills-based hiring?
Skills-based hiring is selecting candidates on evidence of the skills a role requires, rather than on proxies such as degrees, years of experience or previous employers. It has three parts: a short list of must-have skills derived from the actual work, a consistent way to gather evidence of each skill from every candidate, and a decision based on that evidence, scored the same way for everyone.
Does skills-based hiring mean ignoring degrees and experience?
No. It means looking at what a degree or past job produced, not counting it as a credential. A candidate's experience is often the best source of evidence, as long as you ask what they built, owned and fixed rather than how many years they spent. A degree can still matter where the work needs specific knowledge, but test that knowledge directly.
How do you assess skills in engineering interviews?
Use a structured interview, with the same questions and a written scoring guide for every candidate, and a short work sample that mirrors the real job. Ask candidates to walk through one system they built in depth. Each interviewer scores each skill independently with written evidence before the group discusses the candidate.
Why hasn't removing degree requirements changed who gets hired?
Because the rest of the process often stays the same. A 2024 Harvard Business School and Burning Glass Institute study estimated that fewer than 1 in 700 hires in 2023 resulted from dropped degree requirements, and that many companies changed their job ads in name only. If CV screens, interviews and debriefs still reward the old proxies, the outcome doesn't change.
Is skills-based hiring a good fit for AI roles?
Yes. AI roles change quickly, so years of experience with specific tools mean little and many strong engineers came from adjacent fields. Define the skills from the work, such as building retrieval, designing evaluations or running model deployments, and ask for evidence of shipped work. That finds people who have done the job, whatever their title or degree.

Sources

  1. Skills-Based Hiring: The Long Road from Pronouncements to Practice (Sigelman, Fuller and Martin, 2024), Burning Glass Institute and Harvard Business School
  2. Revisiting meta-analytic estimates of validity in personnel selection: Addressing systematic overcorrection for restriction of range (Sackett, Zhang, Berry and Lievens, 2022), Journal of Applied Psychology
  3. The validity and utility of selection methods in personnel psychology (Schmidt and Hunter, 1998), Psychological Bulletin
  4. S.59, Chance to Compete Act of 2024, [Congress.gov](http://congress.gov/)

About the author

Portrait of Lucas Salvaia

Lucas Salvaia

Co-Founder & COO at Tailored Minds

Lucas Salvaia is a co-founder and the COO of Tailored Minds. He works closely with engineers through the whole hiring process, from the first conversation and interview preparation to the offer.

More from LucasLinkedIn

Share this article

Put a whole network on your next hire.

Tell us who you're looking for. Our talent network and recruiters take it from there.

Talent? Join freeRecruiter? Join the workforce