Jim Norris: Memphis, TN

I build AlphaLens at night and on weekends. Fifteen named frameworks. Live prices, SEC filings, current news. Not a scanner. Not an adviser. Not a conversation.

A chat reply is not a research process. The same prompt on the same ticker produces a different essay tomorrow. AlphaLens runs the same named frameworks every time, against data pulled at run time. If the 10-Q is not there, it does not hallucinate the 10-Q.

Every framework runs on the same rule: live data or nothing. Before any framework touches a ticker, it pulls the company's current SEC filings, live price data, and current news, not what an AI model happened to learn during training. If a number isn't available from a live source, the framework says so instead of guessing.

That distinction matters more than it sounds. A stock's price, filings, and news are only useful if they're current, and AI training data has a cutoff date, sometimes a year or more old by the time you're reading this. I built AlphaLens specifically so that gap doesn't become your problem. For a concrete example of why this matters, see the case study on SanDisk's 2025 spinoff, a stock whose entire modern identity postdates most AI models' training data, or the SpaceX post-IPO case study for the same problem applied to a newly public company.

No fabricated price targets. No invented earnings. If the data isn't there, the framework tells you that directly instead of filling the gap with something that sounds plausible. The full research process, the same one built into every framework, is laid out step by step in the Complete Stock Research Checklist.