News & Blog Posts

Why AI often “doesn’t work” in trading—and what changed with transformers. A practical checklist for building production pipelines that beat market-validated baselines.

We talked to quants, PMs, and AI researchers across buy- and sell-side desks to map what actually runs in trading today—organized by time frame and market maturity. On the spectrum from event-driven LOB and imbalance signals (intraday) to legacy ML and macro TS (days→quarters), plus emerging LLM use for signals/portfolio, the winner is whatever beats a market-validated baseline. Our focus: intraday (minutes→hours) using Level-II LOB and finance-native transformer models trained solely on quantitative time-series.

TL;DR: Our latest research shows that quantum agents—learning systems that operate directly on quantum circuits—can teach themselves cornerstone quantum procedures purely from rewards, matching known optimal results. It’s an early but important signal that automated discovery in quantum computing is feasible. At Quantum Signals, we expect trading to progress from today’s algorithmic methods to agentic workflows, and ultimately to quantum-agentic trading. We’re building for the first two—while preparing for the third.

Quantum Signals and Societe General announce their collaboration under the Quantum Pack initiative to develop AI + Quantum trading signals.

At Quantum Signals, we are building AI-powered foundational models for financial time series, focusing on predicting short-term market dynamics such as price movements, volatility, and liquidity. Our platform goes beyond static predictions, allowing financial professionals to customize AI-driven insights by integrating proprietary data, introducing strategic biases, and fine-tuning models to align with specific trading strategies. By leveraging the latest advancements in AI, we aim to democratize access to powerful market forecasting tools and revolutionize how traders and institutions navigate the markets.

Quantum computing offers a fundamentally different approach to computation compared to classical computers that has led to the development of groundbreaking quantum algorithms and cryptographic protocols.

We established an initial benchmark to evaluate mid-price prediction models using Nasdaq order book data, replicating state-of-the-art methodologies from a key research study. By designing a scalable AI pipeline and testing leading models, we validated our approach and set the stage for further experimentation with larger datasets and custom models in AI and quantum-driven trading.

At Quantum Signals, we strongly believe in the transformative power of quantum technology. Our business approach begins with a fully classical AI solution, enabling us to prove our value proposition and establish revenue streams with customers today. Simultaneously, we are dedicated to advancing quantum and quantum-inspired AI methods, ready to integrate them into our pipeline when the technology reaches maturity.

We are thrilled to announce that Quantum Signals has successfully secured a pre-seed funding round with the support of Quantonation.

Quantum Intelligence represents the synergy between quantum technologies and AI and offers novel ways ‍to tackle complex problems and optimize solutions. By integrating quantum-inspired methods and exploring quantum neural networks, we aim to push the boundaries of trading and stay at the forefront of both classical and quantum advancements.

Quantum Signals is enhancing order execution for financial institutions by leveraging AI to predict intraday mid-price movements. By analyzing Level 2 limit order book data and applying advanced generative AI techniques, we aim to optimize trade execution strategies, improving efficiency and minimizing market impact for large orders.