Pelican Alpha: AI-powered equity research for investors

Hey everyone — I'm Ricardo, the founder of Pelican Alpha. Wanted to share the backstory. I'm a CFA charterholder. Spent my career doing small cap equity research at ICONIQ Capital and Osmium Partners (an activist hedge fund). After I left, I tried to start my own fund but couldn't raise enough capital. So I spent two years investing my own money. Nearly 4x'd my personal portfolio. The process worked, but it was brutally slow. One company could eat 40+ hours of my time. Small cap investing is a game of turning over stones — sometimes what you find is obvious, but there are millions of stones to turn over. So I started automating my research process. One Python script became five, five became twenty, and eventually I had a full pipeline that could do in hours what used to take me weeks. That became Pelican Alpha. A few things I'm proud of technically: - The research pipeline is a RAG system — every filing, transcript, and data point is embedded and stored in a vector database so the AI references actual source material - I route different parts of the analysis across 5-6 different LLMs through OpenRouter, picking whichever model is best for each specific task - The AI agent has 35+ tools with a scratchpad that tracks its reasoning chain - The whole thing runs hands-off at 5:30 AM every morning I'll be honest — I'm not a real engineer. I'm a finance guy who taught himself Python. The code is held together with argparse and retry logic. But it works every morning. 270 hedge funds subscribe to the free daily briefing. The paid platform is $50/mo and $100/mo. I'd love feedback on: - The product itself — does the research output feel useful? - Positioning — am I underselling the "hedge fund quality research for $50/mo" angle? - What features would make this more valuable to you? Happy to answer any questions about the build, the finance side, or the AI architecture.