One of the worst things you can do for your career is to become indispensable. That’s how you become the maintenance person. There is nothing “wrong” with being a maintenance person in software or anything else. If you like maintenance or tolerate it, that is great. But there is one problem with being a maintenance-focused engineer in tech: You won’t get promoted. If a service can’t run without you or constantly needs you to keep it up, you will be locked into that job. You become so crucial in that position that your company won’t want to pull you out and stick you somewhere else, even if that is good for your career. This means you won’t get paid more, and you may come to resent the work you are doing. There are multiple ways you can avoid being stuck in maintenance: 1) Always be taking on new tasks. Once you have settled into doing your job well, ask your manager what you can help them with. Continue taking on new tasks and driving larger and larger impact. This stops you from being pigeonholed. I have written about this process multiple times. I call it the “Magic Loop”. 2) Automate with AI If you are in charge of a lot of maintenance, automate as much as possible. This is easier than ever now with AI. This will give you more time to pursue higher impact projects and will make it easier for you to pass off the maintenance to someone else down the line, keeping you from being tethered to it indefinitely. 3) Ruthlessly block your calendar Tasks will take the amount of time you give them (Parkinson’s Law). This means that maintenance will expand to be your entire workday if you let it. Block off the time you are going to use to do your maintenance work and do not let it bleed into time blocked off for new projects or career advancement. 4) Frame your work strategically When reporting on the work you do, don’t say things like “Kept X service running”. That paints you as a maintenance worker. Frame the work in terms of time saved, outages prevented, financial ROI, etc. This will help you be seen as a strategic thinker, not just a technical set of hands. In today’s newsletter, Marcel Tella Amo, Ph.D. has written his guide to avoiding this problem, which he calls “Performance Friction”. He goes in depth about not just how to do what I’ve outlined above, but also how to break your personal points of friction and continue building a successful, fulfilling engineering career. You can read the article here: https://buff.ly/LZgMstv
Writing this with Ethan Evans and Jason P. Yoong pushed me to reflect on how often I had confused "being needed" with "being valued", they feel the same in the moment, but they lead to very different career trajectories. The shift I've seen, especially in tech, is that the engineers and managers who grow fastest are the ones who treat knowledge transfer as a leadership act, not a threat. They build systems and people around them that work without them. What's the hardest part you've had to let go of to move forward? Curious what's coming up for people here.
The indispensability trap is real and it shows up in product too. The PM who becomes the only person who understands a system, a customer relationship, or a process does not become more valuable. They become harder to promote because pulling them out creates a gap nobody else can fill. Building systems and documentation that make your knowledge transferable is not a risk to your career. It is what makes the next opportunity possible.
So true!! I’ve seen a lot of mid-level engineers, myself included a few years ago, get stuck in the “maintenance engineer” trap. You become the go-to person for every ticket, operational issue, and quirky customer escalation. The team depends on you heavily, but you’re often not spending enough time on the kind of high-impact, forward-looking work that actually gets you to the next level. This is a hard place to be in. People around you constantly remind you how important you are to the team and how everything would fall apart without you. Some even directly discuss with you how surprising it is that you haven’t been promoted yet. It’s also a slippery slope for the manager.
If a manager's gut response to a promotion conversation is "but who would handle X," that's a signal worth taking seriously — it’s not a compliment. The work itself isn't the problem; it's how your role has been structured around it, and that's something you can actually change.
Ethan Evans This is so on point. We had a website, it had run for years. Over time, all the original engineers moved on. Except for one. He became indispensable. So much so, he couldn't take PTO. When he went home (to another country) for holidays, he had to take a laptop, and was expected to handle any outages, no matter what else he had going on. It was unsustainable. And yes, he left the business. And they had to completely retool the website because the tech was so old, no-one else had the skills to maintain it. The perfect trifecta: Bad for the employee, bad for the business, bad for the end user.
This is an important point. Career growth usually starts when people stop optimizing only for the work they already know how to do and actively seek out unfamiliar challenges that stretch them.
Automating maintenance with AI usually means doing it faster. The bigger lever is using it to delete the maintenance category entirely so the calendar opens up for new scope rather than just emptier maintenance.
The reframe I'd add: indispensability in a role is a trap; indispensability to the outcome is a ladder. If a system can't run without you, you've quietly negotiated yourself a life sentence in that seat — companies protect what they can't afford to move. The fix isn't doing less good work. It's point 4: "kept X running" and "prevented $2M in outages" describe the same job. Only one of them gets you promoted. Make yourself replaceable on purpose, and frame the work like a strategist, not a set of hands.
A lot of high performers unintentionally build careers around becoming operationally irreplaceable. The stronger long-term move is usually building systems, documenting knowledge, automating repetitive work, and creating leverage beyond yourself. That is what leadership will notice in the end.
Local dependency ≠ Promotable strategic utility The trap isn’t maintenance. It’s becoming the person your manager solves their operational anxiety through. If every messy, fragile, urgent problem lands with you, you become useful. But usefulness to your manager is not the same as promotion value. Promotion starts when your value travels beyond your manager’s immediate pain. That’s when the business starts seeing you differently.