Blog
Research noteBeyond the coding agent: building a research harness from first principlesIntroducing Oh My Quant: a research-native agent harnessResearch noteDeep Learning Autoresearch with GPTHow GPT chooses experiments and interprets their resultsBenchmark reportAutonomous Research on Low Signal-To-Noise DatasetsMeasuring agents’ capabilities in quantitative researchBenchmark reportLong-Horizon Planning in Nonstationary EnvironmentsMeasuring model capabilities to maintain and revise strategies over hundreds of decisions