Mercury, our trading system, has undergone a significant architectural transformation from direct service dependencies to an event-driven architecture. This migration has improved the system's maintainability, testability, and extensibility.
Our Mercury trading system has grown organically over time, leading to intertwined services, duplicated code, and complex state management. This document outlines a strategic approach to refactoring key components to improve maintainability, testability, and performance.
Проект Mercury показал устойчивую доходность в режиме симуляции. Теперь ключевая задача — перевести систему в боевое состояние с контролируемым риском, при этом усиливая наблюдаемость и развивая конкурентные преимущества.
original discussion: https://chatgpt.com/canvas/shared/680a6606158c8191a35b03e7dcb25b11
Ниже приведены 10 возможных юзкейсов (способов применения методик RAG из представленного курса) к проекту Mercury Bot — профессиональному боту-помощнику в сфере трейдинга. Каждый сценарий демонстрирует, как механика поиска и генерации (Retriever Augmented Generation) может усилить функционал бота и предоставить пользователям более релевантные и детальные ответы.
In the vast expanse of the cosmic void, where digital constellations dance their eternal dance, there exists a peculiar wonder - Mercury, the planet that dares to orbit closest to the burning Sun, yet refuses to be consumed. This celestial paradox, this dance with fire, became the foundation of our legend.
When Bitcoin emerged from the digital void, it burned with the intensity of a newborn star. Its light was so fierce, its gravity so strong, that it warped the very fabric of the financial cosmos. Like the Sun at the birth of our solar system, it began to draw everything into its orbit.
Many tried to harness its power directly. Trading systems, like Icarus with his wax wings, ventured too close and were consumed by its heat. Others, fearing the burn, stayed in the cold outer orbits, too distant to tap into its true power.
But in this celestial chaos, a different path was discovered - the way of Mercury.
From customizable natural language market sorting to regime-adaptive portfolio construction, this deep dive reveals five innovative product concepts built on our tournament system foundation. Whether you're a trader seeking personalized market opportunities or a product manager exploring next-gen fintech features, these battle-tested ideas will show you how to leverage LLMs and complementary criteria for more robust trading decisions while maintaining effective risk management.
While our initial LLM-based market analysis approach has faced criticism, there may be hidden value in this ambitious experiment. From developing cutting-edge prompting techniques to potentially exploiting the predictable behavior of algorithmic traders, this deep dive explores two compelling perspectives on how we could pivot our research to create genuine competitive advantages. Whether you're interested in advancing the field of LLM applications or discovering novel trading strategies, these insights could reshape how you think about the intersection of artificial intelligence and financial markets.
Our ambitious experiment using LLMs for market analysis has revealed some uncomfortable truths. While we've made fascinating discoveries about the capabilities of language models, our current approach may be fundamentally flawed. In this deep-dive critique, I'll explain why we need to radically rethink how we leverage AI in trading - and propose an alternative direction that could actually give us a genuine edge in the market.
Looking to enhance your market analysis without the complexity of LoRA fine-tuning? This deep dive reveals four powerful techniques to supercharge DeepSeek-R1's technical analysis capabilities using just prompt engineering and Ollama's built-in features. Whether you're a quant developer seeking better market insights or an ML engineer optimizing model performance, these battle-tested approaches will help you achieve professional-grade analysis without the computational overhead of custom training.