Posted under the CEREVITY organization. Martha Fernandez, LCSW is a Licensed Clinical Social Worker and Co-Founder of CEREVITY, a nationwide private-pay network of independent licensed clinicians treating burnout in engineering leaders, founders and executives.
Every burnout screen you have ever taken asks some version of the same question: has your work started to suffer?
For most of the engineering leads, staff engineers and CTOs CEREVITY treats, the honest answer is no. The sprint closed. The incident got handled. The roadmap shipped. And that answer is exactly why the screen is useless here, because in senior technical roles output is the last thing to degrade and often never degrades at all.
What actually breaks first when a senior engineer burns out?
Judgment breaks first. In the engineering leaders CEREVITY treats, the earliest reliable change is not in velocity or hours, it is in the quality of technical decisions, and it typically shows up two or three quarters before any metric moves.
Your real output at a certain level is not code. It is decisions: what to build, what to deprecate, who to hire, which incident is worth a rewrite and which is worth a comment. None of that is counted anywhere. It has no dashboard. Its quality is only legible in hindsight, by which point it has been absorbed into the general noise of a hard year.
What does degraded technical judgment look like day to day?
CEREVITY clinicians look for four specific changes in how decisions get made. We call the pattern the decision signature.
The option set shrinks.
Ask a well-resourced engineer to scope a hard problem and you get three or four genuinely different approaches, including one that is slightly mad. Ask a depleted one and you get two: the thing we already do, and the obvious alternative. This is usually the earliest signal and the hardest to see from the inside, because a smaller option set feels like decisiveness rather than a narrowed aperture.
Reversible decisions slow down and irreversible ones speed up.
Amazon's 2015 shareholder letter distinguishes one-way doors from two-way doors: what is hard to undo deserves deliberation, what is cheap to reverse should be fast. Under sustained strain this inverts. A feature flag or a config change sits in review for two weeks because it now feels heavy, while a database migration, a framework commitment or a senior hire gets decided fast, at night, with visible relief afterwards.
The relief is worth noticing. It usually means the decision closed an open loop rather than that it was correct.
Commitment to the existing design hardens.
From the outside, conviction and depletion look identical. The difference is what happens when disconfirming information arrives. Someone with capacity argues, then updates within a fortnight. Someone without capacity cannot afford the update, because reopening the question means rebuilding a plan they no longer have the resources to rebuild.
The tell is the disappearance of the "what would have to be true for me to be wrong" conversation.
Delegation collapses exactly when load peaks.
Past a certain point people start pulling work back in rather than pushing it out, and the reason given is always that it is faster to do it myself. Sometimes it is. More often, in a depleted engineer, tolerance ran out rather than time.
What should you measure instead of hours?
CEREVITY opens intake with a two-part question we call the recovery window, and it takes about ninety seconds to answer.
First: how much genuine time away does it now take before you feel like yourself, with nothing owed to anyone before Monday?
Second: once you get there, how long does it hold?
Then answer both again for eighteen months ago, and for three years ago.
The absolute numbers matter far less than the slope. Someone in real trouble will say that three years ago a Saturday did it and it held most of the week; that eighteen months ago it took a long weekend and was gone by Thursday; and that a two-week break last year bought four good days.
This is a clinical heuristic CEREVITY uses in conversation, not a validated instrument, and it diagnoses nothing on its own. What it does is move the conversation off the one question a high performer always wins, which is whether they are still shipping.
Is burnout actually a medical diagnosis?
No. The World Health Organization classifies burnout in ICD-11 as an occupational phenomenon rather than a medical condition, defined as a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed.
That distinction has a practical consequence CEREVITY runs into constantly: burnout by itself is not the thing that gets treated. What gets treated is what has usually grown alongside it, most often a depressive episode, an anxiety disorder, insomnia, or a widening reliance on alcohol to come down at night.
Identifying which of those is actually present, rather than assuming, is the first job of an assessment.
Sleep is the usual starting point, as sequencing rather than wellness, because every cognitive symptom above is amplified by sleep debt. If insomnia has run at least three nights a week for three months and is causing real daytime impairment, it may meet the threshold for chronic insomnia disorder, which a clinician rather than a checklist should determine.
Where it does, the American College of Physicians recommends cognitive behavioral therapy for insomnia as the initial treatment, not sedatives. That is one of the more common starting points in clinical treatment for executive burnout at CEREVITY.
Does treatment mean stepping back from the job?
For most people, no.
The belief that getting treated requires a leave of absence is one of the most durable reasons technical leaders wait, and it does not describe how routine outpatient care works.
CEREVITY runs sessions 8am to 8pm Pacific, seven days a week, in 50-minute, 90-minute and 3-hour formats, by secure video, through a network of independent licensed clinicians. Care is private-pay, which means the therapy generates no insurance claim, no diagnosis code submitted to a carrier and no claims-database entry.
Two things sit outside that: medication is recorded at the pharmacy whoever paid for the therapy, and so is any superbill you choose to file with your own plan. Disclosure obligations that genuinely exist, security clearances and insurance applications among them, turn on what the specific form asks rather than on how care was paid for.
The other thing worth saying plainly is that treatment does not remove the traits that made you good at this. Ambition, conscientiousness and high standards are stable characteristics, and no intervention in psychotherapy makes a person less exacting.
What is treatable is the physiological bill for running those traits in a threat state: the 4am waking, the dread before a planning meeting, the good quarter that produces relief rather than satisfaction.
That distinction is the single most common misconception CEREVITY corrects in clinical care for high-pressure professionals.
One exception to all of the above.
Thoughts of death or of harming yourself are not a point on the burnout spectrum. In the United States, 988 reaches the Suicide and Crisis Lifeline at any hour.
If you have been running on the theory that you must be fine because the work is still getting done, the honest clinical answer is that the work getting done is not evidence of anything. It is the last thing to fail, and in this population it usually fails last on purpose.
Top comments (0)