Personal AI Analyst for $120 a Month: How a Private Trader Earned $180,000 Using Obsidian and Neural Networks
In the world of crypto trading, there are increasingly more examples of technology replacing entire analytical departments. One such case, shared by well-known researcher CyrilXBT, demonstrates how an anonymous trader from China built a fully autonomous market analysis system based on the note-taking platform Obsidian and artificial intelligence. The result is impressive: with a monthly cost of just $120 for the API, approximately $180,000 was earned over six months.
How the "Smart Note-Taker" Works
At the core of the system is a minimalist yet powerful stack: a Mac Mini, an iPhone, and local Obsidian storage. Six automatic pipelines on the N8N platform operate around the clock, collecting into a single repository everything the trader reads, listens to in podcasts, and dictates by voice into a Telegram bot. This is not just data collection, but its intelligent processing.
Every night, the neural network scans about 4,000 linked notes, identifying the strongest correlations between fresh information and already accumulated ideas. N8N acts as the connecting link, bridging data sources with the repository without writing code. Every morning at 6:00 AM, a summary arrives by email: three trading ideas with a confidence rating, an emerging idea of the week, and any note that contradicts an open position. The system only wakes the owner in two cases: when a new note contradicts their current thesis, or when confidence in an idea exceeds 90%.
The Cost and Reality
It is worth emphasizing that all stated profitability figures remain unverified—CyrilXBT has published no supporting data. However, the example itself is extremely telling. The author compares this setup to the work of quantitative funds, where teams of eight people handle the same analytical flow. Here, one person achieved a similar result using a single Mac Mini and a monthly budget of $120 for API fees.
This case is a vivid illustration of a global trend: retail traders are increasingly using combinations of local "note-takers," AI models, and automation platforms. Such systems do not guarantee profit, and loud figures on social media are typically not accompanied by evidence. Nevertheless, the very idea of creating a personal analytical assistant based on one's own knowledge base and neural networks is perhaps one of the most promising directions for retail traders in 2024.
Expert opinion: This approach radically lowers the barrier to entry for obtaining quality analytics. If previously this required a team of analysts and access to Bloomberg, now, essentially, all that is needed is discipline in note-taking and competent automation setup. However, one should not forget that the quality of "garbage in" determines the quality of the result out. The system only amplifies the trader's competencies, but does not replace them.