I tried to make Claude make me money on Algora bounties (data and tool)

A practitioner tested Claude as an autonomous coding agent on Algora open-source bounties across 60 issues with a $20 token budget, attempting to replicate a viral 22M-token, $16.88 payout run. The experiment provides rare empirical data on cost-to-reward ratios for agentic LLM coding loops, directly relevant to anyone evaluating autonomous agent ROI.