The past two weeks have been all about iterating on the trading bot.

Initially, I quickly assembled a prototype using AI, but the code quality plummeted, turning into a tangled mess. Every time I added a new strategy, I had to spend half a day debugging, and compatibility issues kept cropping up. Worse still, when the requirements expanded to tracking multiple token prices simultaneously, the TPS bottleneck of a certain wallet API immediately became apparent, and throughput couldn't keep up.

After some thought, I decided to make a major adjustment. On one hand, I need to preserve the integration logic of the existing wallet API and a certain DEX; on the other hand, I need to redesign the code structure to fully decouple the strategy layer from the data layer. More importantly, I plan to support multiple public chains by using concurrent requests and local caching to bypass the rate limits of individual APIs.

Although this refactoring involves a significant workload, in the long run, it should result in a more stable, easier-to-maintain, and more scalable trading system.