Do Tokens Define Me in an Interview?

Ask what the number is meant to tell you

If you ask me in a job interview how many AI tokens I used last month, I want to know what you hope the number will tell you. Are you asking whether I experiment, whether I can manage a budget or whether I can do the work you need? Those are different questions. I would like the opportunity to answer the one that matters to the job.

I have no objection to discussing how I use AI. A candidate who wants responsibility for these systems should be ready to explain their experience. A large consumption total may reflect useful experimentation. I would still want to hear what the candidate tried, what worked and what they learned before treating it as evidence of competence.

For a text model, tokens are units used to represent and process text, often words or parts of words. Token accounting has practical uses in managing costs and limits. A hiring decision asks something further: what can this person contribute, how do they recognise a mistake and what will they do when the first answer is inadequate?

I would welcome an interview that lets me demonstrate those decisions. Give me an appropriate task, explain the constraints and let me show my choices. Then ask why I accepted one result and rejected another. You will have something more useful to discuss than whether my monthly total looks impressive.

What would count as evidence of ability?

Amazon's engineering memo published on 27 October 2026 warns against turning token consumption into a target, while allowing careful interpretation of input metrics alongside outcomes. Its author, engineering leader Terk Terkowitz, says that when work does not benefit from current tools, “that is a finding, not a failing.”

The memo concerns engineers already at work. It does not establish that a token total predicts applicants’ performance. I draw a question for interviewers from it, not a hiring rule. An employer still has to justify how a measure helps distinguish candidates for the job.

A usage total can prompt a sensible follow-up. An interviewer might ask why a candidate used an assistant extensively for one project and barely touched it for another. The answer could show curiosity, cost awareness or a reasoned decision about suitability. A number becomes more informative when the person explains the work behind it.

My objection is to skipping that explanation. Experimentation deserves a discussion of what was tested and what changed. Efficiency calls for a cost attached to an acceptable result. To assess learning, ask the candidate to explain something they can now do or understand. Token consumption cannot answer all of those questions for you.

The work outside the chat

Anthropic's AI Fluency Index examined 9,830 multi-turn Claude conversations from one week in January 2026. It identified iteration in 85.7% and explicit fact-checking in 8.7%. These are visible chat behaviours, not percentages of people who ever check facts. The classifier assessed 11 of a framework's 24 indicators, leaving other behaviour outside its view.

That measurement limit should influence the way we assess candidates. A person might verify a calculation in a spreadsheet, consult a source in another window or ask an experienced colleague. Their conversation record may omit that work. Equally, a message asking the assistant to double-check does not establish that an independent check happened. I want to examine the evidence, including what took place beyond the chat.

My term Shadow Literacy describes the judgement needed to spot an answer that sounds convincing but is wrong, know enough to verify it and decide what can safely be delegated. An interview can invite a candidate to demonstrate those abilities. Naming them is a starting point; the employer still needs an assessment suited to the role.

Office of Personnel Management guidance on work samples describes assessing activities relevant to the job. It also cautions that the approach suits competencies expected on entry; skills the employer intends to teach may need different treatment. Applying that guidance to an AI-assisted exercise requires design and validation.

A rejected output is worth discussing too. A candidate may have rehearsed their explanation of a successful final product. Discussing an unsuitable draft gives them a chance to show the criteria they used. It can reveal whether they understand why the output failed and whether the correction addresses the cause. The employer would still need to establish whether that discussion reveals something relevant to this particular role.

Explain a decision to leave AI out of a task, too. Perhaps you already had a reliable method, lacked permission to share the material or could not check the tool's answer. I would want to hear how you recognised that limit and what you did instead. That gives a low consumption total some context without asking the interviewer to assume it proves good judgement.

A handover worth examining

Imagine recruiting an analyst who prepares client handovers. Give each candidate the same approved records, permitted tools and usage allowance. In this hypothetical exercise, a client message requests delivery before an agreed review meeting, but an internal schedule puts delivery after it. The dates are visible in both documents. The brief says candidates may ask for clarification.

The interesting decision comes before the polished reply. Does the candidate notice that the two records cannot both describe the agreed sequence? Do they establish who can confirm the deadline, or ask the assistant to pick one and write confidently around it? Acceptable work might be a short explanation of the conflict and a request to the person who owns the schedule. Completing the task responsibly may require leaving the delivery date unconfirmed.

Ask reviewers familiar with the role to agree what a satisfactory answer must establish. Give credit for a justified pause, and explain that asking a question is permitted. Personal access to a paid account should not become an unannounced prerequisite. This is a proposed assessment design, not a validated hiring test; it still needs scrutiny for relevance, consistency and fairness before employers rely on it.

My alternative needs scrutiny too

An interview task can look sensible while measuring the wrong thing. It may reward familiarity with one interface, fluency under time pressure or the assessor's preferred writing style. A candidate who could do the work well with appropriate support may perform poorly in an artificial setting. Those risks remain when we replace a token total with an exercise.

Employers should define the competence being assessed, make relevant adjustments through their established process and use trained reviewers with shared criteria. Avoid giving applicants unpaid assignments that the business intends to use as finished commercial work. Keep the exercise proportionate and explain how the output will be used.

Candidates also face a confidentiality risk when asked to prove past AI use. A chat history can include client information, personal details or material covered by an employer's restrictions. A permitted, edited example and an explanation of the missing context should be acceptable. A recruitment process should not reward someone for disclosing information they were trusted to protect.

Ask the employer to explain the measure

If you are preparing for an interview, ask what the employer wants to learn from its AI questions. Offer an example you are authorised to share, with a clear account of an important choice and the evidence behind it. A low usage total needs an honest explanation of your access and experience. Inventing a higher one would give the employer a false picture of what you can do.

If you design the interview, write down the capability you need before choosing the measure. Separate what the person must know on entry from what you intend to teach. A token-usage question can stay if it opens a relevant discussion; decide which follow-up would distinguish useful experience from expensive trial and error.

Try a proposed work sample with willing colleagues and ask two reviewers to assess it independently against the same criteria. Examine disagreements before discussing names or reputations. A small pilot may reveal an ambiguous instruction or a scoring rule that rewards the wrong thing. It cannot establish predictive validity or fairness across applicant groups, so those questions remain open.

Explain permitted assistance, time expectations, confidentiality and scoring to applicants. Make sure they can demonstrate the skill without paying for access you failed to provide or exposing a previous employer's records. If the interviewer still wants a large consumption number without discussing the work, that tells the candidate something about how performance might be judged after joining.

Questions candidates and interviewers can ask

Can efficient prompting be part of an assessment?

Yes, where cost and usage constraints matter to the role. Define an acceptable result and examine the resources needed to reach it. A short prompt that omits essential evidence can be poor work, while a longer interaction may be justified by a difficult task. Evaluate the trade-off you actually need.

What if the job mainly involves learning new tools?

Give candidates an unfamiliar task with a clear limit, an appropriate introduction and permission to ask questions. Observe how the candidate identifies a gap and checks progress. Keep that exercise separate from testing expertise they were never expected to possess, and assess its suitability before using it to reject applicants.