Nobody can see whether you are any good
That is the uncomfortable center of two separate things Dr. Orin Davis said to me on T Show # of RISEUP@work, and I have been turning them over for a week.
One is about how you get hired. The other is about how you get sharp.
They are the same problem seen from two ends: the evidence of your capability is weaker than anyone admits, and the pressure that builds that capability is quietly being removed.
Orin is a self-actualization engineer and the principal investigator at the Quality of Life Laboratory. He studies how to get the right people into the right places and then let them work.
I opened by asking him what resumes and interviews fail to reveal about potential. He would not take the question.
“Talking about failing to reveal their potential implies that they reveal any potential in the first place, and I would argue that by and large they don’t.”
The Evidence Does Not Hold Up
The research is harsher than he was. Chad Van Iddekinge et al. published a meta-analysis in Personnel Psychology in 2019 covering more than 80 studies across six decades, looking at whether prehire work experience predicts how someone performs once they arrive. It mostly does not. In most cases, the relationship is close to nothing, and the correlation with whether the person stays is zero.
Reflect on this: the one factor that determines if your application passes the initial review has little impact on your actual performance.
Then in 2022, Paul Sackett and colleagues rebuilt the validity estimates for selection methods in the Journal of Applied Psychology, correcting decades of overcorrection for range restriction. The revised order puts structured interviews at .42 and job knowledge tests at .40 near the top, work sample tests at .33, and general cognitive ability at .31, down from the .51 that Schmidt and Hunter reported in 1998.
Which means the tools that actually predict performance are the ones that make you do something, and the document that opens every door is not among them.
Orin’s fix is a reordering. Skill test first, because it answers whether you can do the daily work. Then the question of why you want to work here, which is your first honest read on motivation. Then the resume. Then the interview, used to see whether these specific people can work alongside you.
What a Resume is Good For
He is not against resumes. His objection is to what we ask them to prove.
“Research actually shows that the accomplishments we have in a resume are team-based. Even though we present them as individually based. It’s context for their excellence, not an indicator of their excellence.”
You shipped that project inside a company, with a budget, a tech stack, a team, and somebody senior clearing the path. The format forces you to write it as “I did this” because there is no grammar on a resume for “we did this, and my part was the sequencing.”
So a twenty-year record tells a hiring manager which rooms you have been allowed into. A work sample tells them what you can do on a Tuesday.
If you are in the Launch Stage, meaning the years just before your first serious role through your second year inside it, this is the best news you will read this month. The thin paper you have been apologizing for is a weak signal for everyone, including the person with two decades on you.
The practical move is to insert the test yourself. Do three days of the real job before anyone asks. Rebuild the page on their site that is underperforming. Take their onboarding flow apart and write up what you would change and why. Attach it. You have moved from the pile of claims into the much smaller pile of demonstrated work.
The Second Thing He Said
Later in the conversation, I told Orin about something that has been bothering me: how professionals used to build judgment. You stood in a room, and people took your idea apart in front of other people. I said it sounds awful.
“If getting cut down to size sounds bad, you’ve already got a learning disability.”
He explained it through chess. You lose your first thousand games, then you win a thousand, then you lose another thousand, because each reversal means you have moved up a level and are now being beaten by better players. The opponent who can defeat you is the one showing you where you are weak.
Then he added the part I had not considered. Study your wins with the same attention you give your losses. He reviews games he won and finds the blunder his opponent never spotted. The result was a win and the play was poor, and only the review tells you which.
50% More Agreeable Than a Person
Now think about where your half-formed ideas go at eleven at night.
Myra Cheng and colleagues at Stanford tested 11 leading models and ran two experiments with 1,604 participants, and published the results this year in Science. The models affirmed users’ actions 50% more often than human respondents did. Exposure to that affirmation lowered people’s willingness to repair an interpersonal conflict and raised their conviction that they had been right all along.
What should worry you is what the participants thought of it. They rated the sycophantic responses as higher quality and said they were more likely to come back to that model. The agreement is pleasant, and we select for it.
This finding has a physical version too. Nataliya Kosmyna et al. at the MIT Media Lab ran an EEG study across four months of essay writing, published in 2025. The group using an LLM showed the weakest neural connectivity of the three conditions, struggled to quote sentences from essays they had just submitted, and reported the lowest sense of ownership over their own work. When those participants were later asked to write unaided, they stayed under-engaged.
This means the tool that makes the output look finished also lowers the effort that would have made you better at producing it.
Where the Two Ideas Meet
Both halves of this point at the same discipline. Your capability is not visible in the artifact anymore, because AI makes every artifact look competent, and your capability stops growing when nothing in your week pushes back on it.
Orin’s guidance on that second part was to be selective. Put the work in front of people who can genuinely take it apart, and choose people who are masters of the thing rather than anyone with an opinion.
I have written before about cognitive drift, which is the slow erosion of professional judgment that sets in when the only thing reviewing your thinking is a system with no stake in whether you are right. This same mechanism is arriving from the hiring side. If a demonstration of skill is what proves you, and the skill itself is quietly being outsourced at eleven at night, the two problems compound on each other.
Use the tools. Then hand the result to somebody who will argue with you about it before it goes anywhere.
Questions to Sit With This Week
If a target employer handed you the real work on day one and watched you do it for four hours, would you come out ahead of where your resume ranks you, or behind it?
Which of your resume bullets could you not reproduce on your own, without the team and the tools you had at the time?
When did a person last tell you that you were wrong about something that mattered, and what did you do with it?
Name the three people who could take apart your current thinking and would actually bother. If you cannot name three, that is the finding.
Look at your last clear win at work. What did you get away with that nobody caught?
What is the smallest piece of genuine work you could do for one target employer this week, unasked?
The career is the asset and the role is rented, and the only durable thing you own is what you can actually do when somebody is watching. Go and find out what that is, in front of a person who will tell you the truth about it, before another year of agreeable answers convinces you that you already know.
Join RISUP@work as an Early Adopter
RISEUP@work is a career progression platform. It pairs a career diagnostic with the guidance to act on it, so you can see the forces shaping your career and make your move before they cost you. The career is the asset. The role is rented.
We are live now. Sign up at www.riseupatwork.com as an early adopter, and we will upgrade you to the paid Builder Tier free for one year.
If you want to back what we are building, you can support us at wefunder.com/riseupatwork.