When to Leave Your Job: 9 Honest Signals

You've stopped learning and stopped caring on the same day

This is the single strongest signal. Either alone is survivable — many great careers include quarters where you're heads-down executing on something you already know how to do, and other quarters where the work doesn't excite you but is still teaching you. But when both go flat at once, you've hit the bottom of the value curve. Staying past that point compounds quietly: your tech stops being current, your output gets less interesting, and the version of you that interviews for the next role has fewer fresh stories to tell.

The test: can you name one technical thing you learned in the last 90 days that you'd put on your résumé? If no, and the work also doesn't excite you, the curve has bottomed.

Your manager isn't growing you and won't change

The single highest-leverage thing in a tech career is a manager who actively develops you — gives you stretch projects, advocates for your promotion, defends your work upstream, and tells you the truth about where you're weak. If your current manager doesn't do this and a direct conversation about it produces nothing, you're paying a real career tax to stay. A bad manager doesn't just stall the current role — they cost you the role after this one, because the projects you needed in order to qualify for the next level aren't happening.

The test: ask your manager directly, "what would need to be true for me to be promoted in the next 12 months?" If the answer is vague, contradictory, or arrives without a real plan attached, treat that as your answer.

You've been at the same level for too long, with no real path

Time-at-level is one of the most under-discussed career signals. A senior engineer who's been senior for six years at the same company, with no Staff-track path open and no obvious next step, is in a structural problem — not a personal one. The company has either run out of room for them or doesn't see them clearly. Both are reasons to leave. The market often gives you a level bump just for changing logos, because the new company evaluates you on current work, not on the political history of the last six years.

The test: ask three peers at your level at other companies what their next promotion timeline looks like. If yours is 2x longer and nothing has changed in 12 months, the system is the problem.

The company's strategy has lost the plot

This one is about the company, not you. Watch for the pattern: pivot after pivot with no underlying thesis. A rotating cast of executives. Strategy decks that contradict last quarter's strategy decks. Layoffs that hit good people and protect connected ones. None of this is necessarily fatal — companies pivot, leadership changes — but when the pattern persists for two or more quarters, the steady-state version of the company is going to be smaller, slower, or different in ways that won't be good for your role. The early movers in these situations end up in the best next jobs because they're not competing with everyone else who waited until the announcement.

You're significantly underpaid and the company won't fix it

Underpaid is a high bar — it doesn't mean "I'd like 10% more." It means you've talked to peers at peer companies in your role, looked at credible compensation data for your level and location, and you're meaningfully below the market band. If your manager and skip-level have been told and the response is "we'll look at it next cycle" twice in a row, the market will fix it for you. Companies that won't fix known comp gaps for proven performers are signaling something about their stance toward you specifically — and that signal almost always shows up later in promo cycles too.

The team's standards have dropped and aren't coming back

Working on a team with high standards is what makes engineers better over time. When the bar drops — sloppy reviews, missed deadlines that go unaddressed, "good enough" becoming the default — your own standards drift to match. This is one of the most invisible and dangerous career costs in tech. The version of you after three years on a low-standards team will be worse on every dimension that matters: code quality, judgment, sense of what shipping really means. If standards have visibly dropped and leadership isn't trying to bring them back, the team is hardening into a long-term plateau.

You're being asked to do work that violates your values

This is the rarest signal but the most clear-cut when it shows up. Being asked to ship something you believe is misleading to users, being asked to participate in a layoff process you think is unjust, being told to ignore a safety or security issue because shipping matters more — these are signals you don't try to renegotiate with. You raise it once at the right level, document it, and start looking. Compromising on this in tech has a long memory: it shapes how you talk about yourself in the next interview, what kinds of companies you can credibly tell a story about working at, and how you feel about your own career when you're 50.

Sunday-night dread has become Sunday-night certainty

Most jobs have Sunday-night moments — that wave of dread when you remember Monday morning. Healthy careers have them occasionally, when something hard is on the calendar. The signal isn't the dread itself; it's the consistency. If every Sunday is the same dread and there's no specific thing causing it, your body is telling you something your brain has been arguing with for months. Bodies are more honest than narratives. Trust the pattern.

The test: have you used a sick day or PTO day to avoid a specific recurring meeting? More than once? That's a real signal.

The team you respected has quietly left

Watch the exits. The engineers you most respected, the manager who hired you, the PM who made cross-functional work feel possible — if a meaningful number of them have left in the last 12 months, that's a leading indicator the company didn't catch. The people closest to the work usually leave first when something structural breaks. By the time the official announcement comes, the people who could have taught you the most are already at the next place. You can either be among them or among the ones who realize a year later that the culture you joined no longer exists.

You had a bad month

A failed launch, a brutal review cycle, a conflict with a peer, a project that died — any of these can produce a week of "I'm done." That's not the quit signal. The signal is when the bad month becomes a bad pattern across three or four months and nothing's been done to address it. A single bad month inside an otherwise healthy job is almost always survivable, and acting on it usually produces a worse next job rather than a better one. The right move: wait 30 days, talk to someone outside the situation, then re-assess.

You're bored, but you haven't asked for new scope

Boredom is real, but it's also one of the most fixable problems inside a current job. Before you treat boredom as a quit signal, run a four-week experiment: tell your manager you want a bigger or harder problem, propose two specific ones, and see what happens. If your manager engages and the work materializes, you've solved the problem cheaper than a job change. If nothing happens, that's its own signal — but it's now Signal 02 (manager isn't growing you), not generic boredom.

You're tired

Burnout often looks identical to "I need a new job," and it isn't. The fastest career mistake is to quit-and-start at a new company while still depleted — you walk in unable to ramp, hit the new-job honeymoon-then-crash earlier than usual, and end up six months later in the same place but with a one-year tenure that's harder to explain. If you're physically and emotionally exhausted, the real question isn't "should I leave?" — it's "can I take a week off and re-ask the question?" Recovery before deciding is almost always a better path than deciding while exhausted.