Erik Lindqvist

Answered How much does temperature actually matter for code generation?

Benchmarked on 150 tasks: temperature 0 and 0.3 were within noise on correctness. What did differ was diversity across runs — at 0 you get the same wrong answer five times, at 0.3 one of five attempts is often right. So it depends on your loop. Single-shot into production: 0. Generating candidates you filter with tests: higher, and take the one that passes.

Shared a bide Turn Messy Meeting Notes Into Decisions, Owners and Deadlines

Paste raw meeting notes and get decisions, actions with owners and dates, open questions and a ready-to-send recap, without invented commitments.

Shared a bide Compare two coding agents' attempts at the same task and pick one

Score two AI coding agents' solutions to the same task on a weighted rubric, pick one with reasons, and list what to cherry-pick from the loser.