Jeff Ma(@18jeffreyma):super timely work, and a really great explanation for why models might feel "benchmaxxed"! SWE evals definitely feel extremely Markovian, which lends really well to compaction (since your codebase is essentially agent memory) real-world agent use is clearly multi-turn and partially-observable (uncovering requirements once you see your own spec assumptions challenged and overturned)

2026.10.30 17:56

super timely work, and a really great explanation for why models might feel "benchmaxxed"! SWE evals definitely feel extremely Markovian, which lends really well to compaction (since your codebase is essentially agent memory) real-world agent use is clearly multi-turn and partially-observable (uncovering requirements once you see your own spec assumptions challenged and overturned)