Being a good engineer kinda sucks

Good engineering can feel miserable when speed and competence collide with team insecurity—and the long-term fix isn’t technical, it’s social. A personal career arc at Twitch shows how being “the best engineer” on the wrong team can trigger political backlash, HR involvement, and a cascade of stress, even when the work is objectively correct. The core lesson: leverage and impact matter, but so does the environment that turns competence into support rather than threat.

The rant starts with a career reality check: engineers should ask how replaceable they are—often through “bus factor.” If disappearing barely hurts the company, promotions and bargaining power will be limited. If disappearing would cripple delivery, the engineer can become trapped maintaining systems indefinitely. That framing sets up the speaker’s experience of being essential on one team, then later discovering how quickly that “essential” status can become a liability.

At Twitch, the speaker’s early career involved repeated reorgs and unstable assignments, including a video-on-demand platform team that was cut soon after joining. A move to the safety team changed everything. Within weeks, a manager pushed back on the speaker’s level and pay, effectively advocating for a promotion and forcing honest self-assessment. With that support, the speaker gained confidence to propose architectural and tooling improvements, negotiate raises, and hire additional talent. The team also created cross-team “swaps” to unblock roadmaps—sending engineers temporarily to help other groups ship safety features faster. In that setting, shipping faster improved both outcomes and relationships.

Later, a team swap to the creator org exposed a harsher dynamic. The speaker left a safety seat that couldn’t be refilled by simply hiring another mid-level engineer; speed and delivery patterns were hard to replace. In the new org, competence drew suspicion. The speaker described being blocked for months on tickets, then taking them on and finishing them in hours—only to face hostility. A manager monitored the speaker’s calendar and escalated concerns about meetings with higher-ups. The tension escalated during a hackathon where the speaker rebuilt the mobile app from scratch, won, and still received an HR warning after leadership was influenced by the mobile team’s complaints.

The most vivid example came from a dashboard change. After a long discussion with a product manager about why syncing layouts across devices would break usability (percentage-based layouts don’t translate across aspect ratios), the proposal returned unchanged weeks later. The speaker exploded in the meeting, arguing with data from an aspect-ratio experiment and user statistics. Even when correctness was supported by evidence, the political cost was real.

The speaker then ties these workplace conflicts to a broader principle: stop doing work that doesn’t benefit one’s role and leverage—an advice they failed to follow. When the environment punishes speaking up, “doing the right thing” can damage career outcomes. The closing argument shifts from grievance to strategy: engineers need motivated allies. Competence matters, but trust networks matter more. The speaker credits mentors, friends, and defenders for sustaining growth and ultimately enabling a career outside Twitch. The practical takeaway is to find teams where improvement is rewarded, and if they don’t exist, move—because otherwise the only outcomes are burnout, suppression, or relocation to a healthier environment.