How do you become known as a good engineer?

Open original

The author, an established engineer who gets hundreds of resumes a month, argues that engineering reputation comes from doing genuinely hard, self-directed work and then talking about it in front of a small, relevant community. He lays out concrete ways to do that: attending niche meetups, writing up projects, and submitting them to forums like Hacker News.

  • Pick one community you respect technically and aim to contribute to it, not just to get something from it; impressing one good audience usually carries over to others.
  • Good work means weeks or months of effort on something that teaches you what most peers haven't done, not a resume-padding side project or a standard assignment. Experts can usually tell the difference.
  • Showing up matters more than people expect: go to small niche meetups of 10-20 people rather than 100+ person events, and write up projects in enough depth to show your reasoning and the problems you hit along the way.

The author, an established software engineer who gets hundreds of resumes a month, starts from a conversation with a younger engineer who wanted to know how to get known as good. His core answer: pick one community you admire technically (the audience matters less than you'd think, since impressing one tends to carry over to others), then focus on what you can contribute to it rather than what it can do for you. He cites Paul Graham's point that self-directed projects teach far faster than coursework, and he's blunt that the usual resume tricks, like a fine-tuned YOLO model or a marginal paper squeezed into NeurIPS, fool few people who actually know the field. A good project, in his framing, takes weeks or months, teaches something most peers haven't done, and is something you genuinely find interesting. At a job, he suggests explicitly asking your manager for interesting work, then working hard once you get it. On the communication side, he argues that doing good work isn't enough, since plenty of strong engineers get overlooked. His first recommendation is showing up in person. Skip the 100-plus-person events and go to small niche meetups of 10 to 20 people, like a Rust group or a Discord meetup in San Francisco, not the sponsor-heavy ML showcases. He recalls one AI meetup with just three or four people where everyone spent the evening demoing projects, which he calls among the best he attended. Outside tech hubs, a Discord or open-source contributions can stand in, but only if you contribute real work and stick around. Writing comes next: he recommends publishing posts about the reasoning and dead ends behind a project, then submitting them to Hacker News, relevant subreddits, or lobste.rs, and sending them directly to mentors who can boost them. The piece closes its visible portion by arguing that strong writing is itself a signal of engineering skill, pointing to Ben Kuhn's "In defense of blub studies" as an example of a post that shows depth through synthesis across many projects. The excerpt cuts off mid-sentence in its discussion of how great engineers reduce complicated problems, so the rest of the argument isn't available here.

Ask about this article