David Len
The feeds are designed to forget.
Twitter surfaces what's new. LinkedIn surfaces what's active. GitHub surfaces what's been committed to recently. Every platform has made a structural bet that recency equals relevance, and their algorithms execute that bet relentlessly.
This is fine for news. It's a problem for evaluating people.
Because the most interesting things someone has written are often years old. The insight that held up. The take that looked wrong at the time and turned out to be right. The technical post they wrote for themselves that nobody read but that shows exactly how they think through a hard problem. None of this floats to the surface in a recency-biased feed.
The result is a kind of epistemic amnesia in hiring. You see what someone posted last month. You almost never see what someone posted four years ago that aged well.
There's also a commercial pressure that makes this worse. As platforms have become more crowded and more commercial, the identities competing for attention have tilted toward brands, products, and professional personas. The original use case — people sharing what they think and find interesting — gets increasingly drowned out. The personal, the educational, the genuinely curious: all of it competes poorly against optimized content from accounts with budgets and strategies.
What gets lost is the most honest signal: someone doing the work of thinking in public, over a long period, with no particular audience in mind.
CloudScout doesn't just read the recent stuff. It reads everything it can find, weights for consistency over time, and surfaces the thinking that held up rather than the content that performed well. A post with four likes from 2022 that turned out to be correct is worth more to us than a viral thread from last week.
The platforms forgot it. That's not a flaw we have to inherit.