Ok I’ll rephrase: We software engineers tried to buy/sell houses using algorithms, and failed to the tune of $550m. I think we collectively would have similar success (or lack thereof) as actual real estate agents.