The illegible nature of software development talent

Metrics, Interviews, and “Legibility” of Talent

• Many commenters say common hiring signals (LeetCode, algorithm drills, buzzword stacks, GitHub activity) correlate poorly with who actually delivers value.
• Coding interviews are likened to the “streetlight effect”: we rigorously measure what’s easy (toy problems) instead of what matters (architecture, judgment, collaboration).
• Work trials are seen as more accurate but impractical and biased toward young/unemployed candidates.
• Several people note that even for “staff+” roles, companies default to the same commodity interview loop that’s detached from day‑to‑day work.

What Good Developers Look Like (and Don’t)

• Anecdotes highlight top engineers who look very different: self‑taught hackers vs academic high‑achievers; obsessive side‑project people vs strict 9‑to‑5 workers.
• There’s a strong distinction drawn between “brilliant programmers” and people who reliably deliver results in a team; these sets only partially overlap.
• Some argue exceptional devs are often underused or misdirected (e.g., pulled from hard systemic problems to generic urgent tasks), turning “10x” into “0.5x”.
• Others warn against mythologizing “stealth 10x” types and downplaying the role of HR and structured processes.

Side Projects, Passion, and Expectations

• Multiple commenters push back on the idea that real talent requires coding outside work. Many stop once careers and families mature, or shift hobbies away from screens.
• Interview questions about off‑hours projects are debated: some see them as harmless chances to surface extra experience; others see them as inappropriate, coercive, or even legally risky.

Environment, Management, and Recognition

• People report huge performance differences based on environment: private offices vs open plans, autonomy vs top‑down control, space to fix real problems vs being forced to “look busy.”
• Preventative work (avoiding crises, building robust systems) is described as high‑value yet mostly invisible in reviews and promotion processes, which favor visible heroics and “playing the game.”

“Illegible” vs “Invisible”

• Several comments dissect the title term: “illegible” (from Seeing Like a State) is defended as meaning “exists, but can’t be parsed by large organizations’ metrics,” as opposed to truly “invisible.”
• Some find the word trendy or overwrought, preferring “intangible” or “inscrutable,” but others argue it captures the mismatch between real value and what institutions can easily read.

Risk, Markets, and Alternatives

• Some see hiring as a “Market for Lemons”: noisy signals push high‑quality people out of broad markets into trusted sub‑networks, referrals, or long‑term gigs.
• Startups are framed as “talent arbitrage”: founders poach undervalued but proven colleagues.
• A provocative suggestion to use AI to score individual workers on their output is widely criticized as impractical, ethically dubious, and blind to crucial but hard‑to‑quantify skills.

Networking and “No Presence” Careers

• Several strong engineers report having no online profile, minimal public code, and careers built entirely on word‑of‑mouth and ex‑colleagues.
• Commenters stress the importance of maintaining professional relationships as a hedge against being illegible to resume filters and external metrics.