Most retail traders rely on emotion and noisy headlines. Engineered by a mathematics professor, StokeStox levels the playing field. Our dual-engine platform combines custom machine learning (tested with strict out-of-sample data) with Gemini 3.0 to generate clear 0-100 stock scores. Stop guessing and chat directly with institutional-grade data. Features include multi-portfolio sync, AI explanations, and walk-forward backtested signals accounting for realistic slippage.
No reviews yetBe the first to leave a review for StokeStox
Maker
📌
Hi Product Hunt! 👋
I’m Michael. I’m a mathematics professor, a retail trader of 20+ years, and the solo developer behind StokeStox.
For two decades, I traded my own accounts and kept hitting the same wall. Despite my math background, I was frustrated by retail tools that felt like black boxes—either noisy sentiment trackers chasing social media hype or technical analysis that didn't hold up out-of-sample.
I knew institutional-grade math existed—I teach it—but there was no clean way to apply strict quantitative machine learning to my own daily trading. I actually lost followers on X when I shifted from "entertainment" to this deep math approach. But that's okay—I'm looking for the investors who care about Alpha, not hype.
🧠 Under the Hood:
I built custom ensemble ML models that calculate a strict 0-100 score for thousands of stocks based on years of financial data, backtested with realistic slippage. But because data without context is just noise, I’ve integrated Google’s latest Gemini AI directly into the engine. This means you aren’t just staring at a chart; you can actually chat with your portfolio to understand the "why" behind the movement through Explainable AI (xAI).
🛠️ The Stack:
Built solo using React/Node.js and Firebase for a snappy, real-time frontend, with Python (Pandas/Scikit-learn) handling the heavy ML pipelines and institutional feeds from FMP.
🎯 Why I’m here:
I didn't build this to stay on my local machine. I want to see if it holds up in the wild. I’m looking for data-driven traders and developers to battle-test this engine. Go score your own portfolios, run the AI Analyst, and honestly... try to break the system.
🎁 PH Special:
The scanner and basic scoring are free. If you want to push the system to its limits unlocking 250,000 AI tokens/month (upto 350 quantitative assistant queries) and full factor breakdowns, use the code FOUNDER40 at checkout for 40% off lifetime Pro access as an early tester.
I’ll be here all day to geek out over quant math, trading, and Node.js architecture. Let me know what you think!