Unconventional Hiring Ideas for Engineers | Generated by AI
Question: What are some unpopular or unconventional ideas about hiring engineers?
Answer:
Here are several genuinely contrarian or unpopular views on hiring engineers that go against mainstream tech-industry advice (résumés, LeetCode, “culture fit” interviews, etc.):
-
Ignore résumés and pedigrees almost entirely
The best predictor of future engineering performance is not where someone went to school or which FAANG company is on their CV. Many top performers have unremarkable or messy résumés. A surprising number of companies (e.g., some hedge funds and elite startups) throw away the résumé before the first phone call and start with a fresh work trial or technical conversation. -
Hire for spike, not balance
Conventional wisdom says “look for well-rounded engineers.” The unpopular view: deliberately hire extreme specialists (e.g., someone who is 10× on low-level systems but mediocre at web frontend) and pair them with complementary spikes. Balanced mediocrity is more dangerous than lopsided brilliance. -
Never use algorithmic whiteboard/LeetCode interviews
Widely practiced, yet many high-performing teams (Basecamp, Stripe in its early days, Jane Street for non-quants, etc.) either banned them entirely or treat them as a very weak signal. Real-world coding trials, take-home projects that resemble actual work, or pair-programming on the existing codebase are stronger signals and less gameable. -
Prefer “has shipped and maintained ugly but working software for years” over “clean-code purist”
Engineers who have lived with their own technical debt for 3–5 years tend to make better long-term decisions than those who have only worked on greenfield or short-lived projects. -
Hire older engineers (40–50+) aggressively
Ageism is rampant in tech, yet engineers in their 40s and 50s often have dramatically higher productivity, better judgment, and lower ego than 25-year-olds—if you can find the ones who never stopped coding. Companies that figured this out (e.g., some finance and embedded firms) keep it quiet because the supply is limited. -
Don’t hire for “culture fit”; hire for “culture addition” or even culture friction
The most innovative teams often have a few abrasive, high-talent people who challenge the status quo. Homogeneous “no jerks” policies can accidentally filter out the exact people who drive disproportionate impact. -
Pay people to reject your offer
Some companies (e.g., certain prop trading firms) give candidates $5k–$10k just for finishing the interview process, even if they say no. Result: candidates are brutally honest in feedback, and the acceptance rate on offers skyrockets because only truly interested people remain. -
Use reference checks as the primary signal, not a formality
Most companies treat back-channel references as a check-box. The contrarian approach: spend 3–5× more time on references than on interviews. A single strong reference from someone you trust beats ten interview rounds of interviewing. -
Hire people who failed spectacularly (and can talk about it intelligently)
Founders or engineers who have been part of big failures (bankrupt startups, cancelled projects) often have scar tissue that is more valuable than another flawless Google tenure. -
Trial periods > offers
Instead of extending an offer and hoping, bring people in for a paid 2–8 week contract on real work. Conversion rates from strong trials are 90%+, and both sides know it’s a fit. (Common in Europe, still rare in the US.)
These ideas are unpopular because they are higher-friction, legally trickier in some jurisdictions, or offend common fairness intuitions—but the teams that adopt even a few of them often end up with dramatically stronger engineering organizations.
References:
- The Year Without Pants – Scott Berkun (on Basecamp’s no-LeetCode hiring)
- Hiring and the Market for Lemons – Dan Luu
- Don’t Hire the Best – Keith Rabois (YouTube talk)
- Painless Reference Checking – Rands in Repose