Lazy algorithms for winning, People Edition

Here is a people hack that’s very lazy. We use it for hiring engineers.

State your project. See who responds. Filter very lightly. Hire 30% of
them. Start with a trial. Increase hours and difficulty. At any stage, any
party terminates.

Negatives:
– you can only hire people who want to do trials or work freelance
– such people may be transient (eg not “career Googlers”)

Non-negatives:
– they will do annoying or poor work or damage your project — roll back in
got
– they require tons of oversight — point them at your bug list/project
cards
– you can’t verify their work — use a continuous deployment dev environment
– their code quality – use them to check each others checkins Mechanical
Turk style

Positives:
– zero management time hiring
– 85% lower cost
– ability to reduce cost structure immediately
– ability to hire quickly

The code as craft people obviously hate this. But too bad for them. Results
matter not theory.

Note irony: their invention of devops/continuous deployment makes it all
possible.