Field notes on AI coding agents — what they cost, where they break, what shipped. Every published figure (prices, percentages, multiples, token counts and durations) as JSON and CSV, each row carrying the sentence it came from.
5 of the 24 write-ups here are tagged AI Implementation. Every figure quoted below is in the figures table with the sentence it came from.
A specialized code review agent beat Claude Code on accuracy across 200 real pull requests and 50 open-source repositories while burning about one-ninth the tokens.
Microsoft’s evaluation of Kimi K3 landed on a number that should change how you read a pricing page: about 60 percent of the cost difference between models comes from the thinking depth a task requ…
The agent processes 40-plus podcast channels overnight, transcribed and summarized, ready to read by morning.
“Choosing the right GPT-5.6 model for your business is more about avoiding cost overruns than just picking the cheapest option.”
Klarna reported $4 million a year in savings and a 99.96 percent conversation engagement rate, the kind of pair of numbers that ends an internal debate before it starts.