Ron B.
I’m getting increasingly frustrated…
I’m getting increasingly frustrated with Composer’s Discover / AI usage limits. Tonight I spent a serious research session building and testing a strategy family. This was not casual clicking around. It took about 10–12 attempts to work through the logic: Top 1 vs Top 2, SPY MA200 defense, removing a ticker, Top 3, shorter lookback, different defensive logic, and finally a fast-defense version that actually looked promising. That is how strategy research works. Most of the early attempts are supposed to fail. The failed branches are what teach you where the real signal is. The problem is that by the time I finally found the promising candidate, I hit the Discover / AI limit and could not continue evaluating the symphony until the reset. That feels completely backwards. The limit gets consumed during the messy exploration phase, then cuts the user off right when the work becomes valuable: checking the final logic, reviewing holdings, comparing the surviving version, and deciding whether the strategy is actually worth watching or buying. I understand having limits. What I don’t understand is blocking review of symphonies that were just created because the failed research attempts used up the allowance. A better system would separate building from reviewing. If I hit a creation limit, fine — stop me from building more. But I should still be able to evaluate, compare, and inspect the results of what I already created. At minimum, Composer should show a clear remaining-use counter and give serious users a way to finish the current research session. Right now the limit interrupts the process at the exact moment the research starts to matter. For a platform built around strategy discovery, that is a major workflow problem.