Sophistia lets you type any factor in plain English - e.g., “AI datacenter exposure”, “high switching costs”, “mission-critical supplier”, “tariff sensitivity”, “recurring-revenue strength”, “rare earths value chain”, “Trump trade war beneficiaries”, “AI-disruption risk”, etc. You can describe in detail what the factor should evaluate and choose the scoring scale.
The system scores thousands of companies from 0–10 on each factor using LLMs with structured context. Companies with the highest total scores rise to the top of your custom thematic watchlist. It’s essentially a factor-model stock screener for retail investors: you define the narrative, factors, characteristics, or thesis, and the tool finds the companies that actually match it.
I built this after realizing how difficult it is for retail investors to identify companies that benefit (or suffer) from a trend without already knowing the entire universe of stocks. Hedge funds solve this by having the resources to track all the stocks; retail investors don't. This is my attempt to bridge that gap.
This is not a tool for deep, company-specific analysis but rather a way to surface the companies that exhibit the characteristics you’re looking for and may deserve a closer look.
This is very much a v1: right now it only covers SEC companies and the contextual data is still limited. Feedback from anyone familiar with active trading, screening tools, or factor models would be extremely helpful.
This is a completely solo-founded project that I built myself.