Hello, Product Hunt. Can you see me this time?

I joined Product Hunt over six months ago and have been turning up every day, weekends included. In times like these, people are still making things with care. I thought the least I could do was leave a comment and an upvote. Besides, AI doesn’t take Sundays off, and I didn’t feel qualified to declare a day off on behalf of the entire industry. Then one day, starting with one particular comment, everything went quiet. No replies. No echo. A very thorough sort of silence. I opened a browser where I wasn’t logged in and found that my comments weren’t visible at all. I’d thought I was sitting in a public square, talking to people. Perhaps I was in a room for one, furnished to look like a public square. “Why isn’t anyone replying?” soon became three much older questions: Who am I? Where am I? What am I doing? So here I am in the introductions section, which looks like somewhere you can start again—the spawn lobby of a game. Hello, everyone. Before we go any further: did I spawn on the same map as you this time?

While there’s a chance someone can see me, I’d like to introduce StratCraft, the AI-native platform for end-to-end quantitative research I’m building. I want people without a background in quant trading to be able to start with an idea and gradually build their own trading system, helping it grow through testing, failure, and revision. The LLM can let its imagination wander; the code has to follow the rules, and the experiments have to leave evidence that can be reproduced. I think of this accumulation as layers of rock. The Earth keeps its years in stone; a market keeps its history in candlesticks. An AI tool can keep something of each LLM conversation too: code, and a traceable record of how it came about. The code goes into a backtest. The backtest produces data. That data, together with earlier conversations, becomes context for the next LLM call. The next thought has a foundation. I want each call to leave something in these layers: trading algorithms as granite, factors as basalt, machine learning models as obsidian. Work that fails validation leaves a record too, for the next call to consult. The LLM is a visitor passing through. The code, the data, and the history stay, slowly becoming the continent beneath its feet when it returns. Whenever I look at those invisible comments, I find myself thinking about the rules people have always lived with: natural rules, social rules, some plainly visible, others discovered by walking into them. The internet and AI give some of them a shape we can begin to make out. I hope StratCraft will put tools for studying them into ordinary people’s hands, so they can test their judgments and find their own ways of dealing with the market. As for me, I’m working on a question of my own: how does a person introduce himself on PH and actually appear?