Skip to content

Stalled DevOps Career? Diagnose the Unexamined Close Calls

Published November 21, 2026 by Beena Yadav

Picture an engineer with four years of DevOps experience who keeps getting the same feedback at review time: solid technical skills, but growth seems to have levelled off. Colleagues with similar tenure are being handed larger systems and mentoring duties. This engineer is perfectly capable. The problem is hiding somewhere less obvious, and a diagnosis usually points to the same culprit: a pile of close calls that taught nothing because nobody looked at them. Below, a problem-and-solution walk through how that happens and how to reverse it.

The symptom: busy, competent and stuck

Our imagined engineer stays on top of tickets, deploys without drama and responds quickly during on-call weeks. The surface is calm. Yet over a year the engineer has had a dozen scary moments: a rollout paused at the last second, a database change that almost ran against the wrong host, a monitoring gap discovered only by accident. Each moment ended well, each was forgotten by the next stand-up, and each would have made a fine lesson.

Careers in operations tend to deepen through judgment, and judgment is built from reflected experience. An engineer who reflects on failures only after they become outages gets one or two lessons a year, at high cost. An engineer who reflects on close calls gets dozens, at no cost. The gap compounds over time, and managers notice it as “range” or “instincts” without knowing where it came from.

The diagnosis: three habits that let lessons slip away

Habit one: treating a good outcome as proof of a good process

When a deployment succeeds, the brain files it under “fine”. Yet a successful result built on a lucky catch is not a sound process. Asking “would this have worked if I had been five minutes later?” separates solid practice from fortunate timing.

Habit two: moving on because the next task is waiting

Pressure to ship makes pausing feel like a luxury. But a two-minute note while the memory is fresh is far cheaper than rediscovering the same weakness months later. Busy people need the capture step to be tiny, not absent.

Habit three: keeping lessons private

Even engineers who do reflect often keep it to themselves, out of embarrassment or habit. The result is that the team repeats the same slips in parallel. Sharing a scare in a stand-up costs thirty seconds and tells every colleague what to watch for.

The treatment: a four-part routine

Capture

Keep a running note with a single line per event: what almost happened and what stopped it. Add the date. Do not edit or judge; just write.

Classify

Every couple of weeks, label each line with the skill area it touches: environment awareness, change planning, observability, communication, access control. Over a few months, the distribution tells you which skill needs work.

Practise

Pick the most frequent category and give it deliberate attention. If environment mix-ups dominate, spend an afternoon building a safer setup, with distinct prompts, separate credentials and a pre-flight check script. If observability gaps dominate, review the alerts on one service and find what you would not see during a real failure.

Share

Present one learning per month to the team, framed as a pattern and not a confession. This builds the reputation that senior roles depend on: someone who notices weaknesses and fixes them proactively.

How this connects to the usual career roadmap

Career guides generally list skills to acquire: cloud platforms, automation, containers, monitoring. These matter. But advancement from junior to senior usually hinges on something those lists underplay, which is the ability to anticipate how a system will fail. The overview of how a DevOps career typically progresses from junior to senior level describes the stages, and the near-miss habit speeds up the transition between them because it feeds anticipation directly.

Likewise, anyone planning a long-term skills programme will find the near-miss list a useful input to a career roadmap spanning skills, projects and growth. Instead of guessing what to learn next, you can let your own close calls nominate the topics.

A concrete before-and-after

Return to our engineer. In the first quarter after adopting the routine, the note collects fourteen lines. Classification shows that six involve environment confusion and four involve alerts that did not fire when expected. The engineer builds a pre-flight script that prints the target environment and asks for confirmation, and spends one afternoon reviewing alerts on the service with the most gaps.

By the third quarter, environment scares have dropped to one. The engineer presents the pre-flight script to the team, and two colleagues adopt it. At the next review, the manager mentions initiative and “a gift for spotting risks early”. Nothing about the engineer’s technical ability changed. What changed was the use of experience already on hand.

Ruling out other causes

A fair diagnosis considers alternatives. Stalled growth can also come from a role with no stretch work, a team with no mentors, or an organisation that rewards tenure over contribution. If you have been capturing and sharing close calls for six months and nothing shifts, look at those factors instead. The routine above is not a cure-all; it is the cheapest fix for the most common cause, and it makes the remaining causes easier to see since your notes already show what you have been absorbing.

What a manager can do

Managers who want to help can make three small moves without adding process.

  • Ask in one-to-ones, “what almost went wrong this month?”, and listen without correcting.
  • React to each answer with interest and a question about what would make the catch dependable.
  • Give people room to fix the thing they found, even if it is slightly outside the current sprint.

The effect is double. People learn that spotting weaknesses is valued, and the team gains steady, low-drama improvements. In our experience of how such conversations tend to go, the engineers who become most trusted are those who bring problems early, and managers get that behaviour by rewarding it when it first appears.

A second short story: the quiet backup

Consider a different engineer, three years in, who was asked to restore a small database to a test server for a colleague. During the restore the engineer noticed that the newest backup was nine days old, not one day old as everyone assumed. The scheduled backup job had been failing silently after a credentials change. No data had been lost because the production database was healthy. The engineer fixed the credential, added an alert for stale backups and wrote two lines about it in the running note.

The event never reached a manager. But when the same engineer was later asked to lead the effort to harden backup practices, the notes provided specific, credible material, and the proposal was approved quickly. A near miss, quietly recorded, became the foundation of a larger project and a visible step up in responsibility.

Side effects worth expecting

People who adopt this habit often notice smaller benefits along the way. Interviews become easier, because they have specific stories to tell. Handovers improve, because they have a list of known hazards to pass on. Confidence rises, not because mistakes vanish, but because each one is now treated as information.

Warning signs that the routine is failing

  • The note has not been touched for three weeks. The capture step is too heavy; shrink it.
  • Entries are all vague, such as “something felt wrong”. Add the prompt “what exactly did I see?”.
  • Nothing ever changes after an entry. Pick one entry per fortnight and fix it.
  • You feel guilty reading the list. Reframe it as a record of catches, not failures.

A wider perspective

This is not a new idea. Fields that depend on reliability, such as aviation and online service operations, have long used close calls as teaching material. A website on that subject, SRE School, has a blog-style front page of short reliability articles and sits in that tradition. Engineers on any career path can borrow the underlying mindset without taking on any particular framework.

Pulling it together

The diagnosis in one line: stalled growth often has less to do with missing skills than with unprocessed experience. The remedy is as plain as the problem. Write down the close calls, look for patterns, practise the weak spot and share what you learn. It takes minutes per week, costs nothing in tools and compounds quietly. Twelve months on, the notes double as a self-made curriculum and as the strongest proof of judgment you could hand a hiring manager.

Start tonight. Open a blank note, think back over the past week, and write down the last moment that made you exhale with relief. That line is the first page of a better career story.