A national U.S. Stock Exchange transformed a manual SEC reporting process into a specialist AI model, with the goal of achieving complete accuracy as well as significant combined savings and revenue uplift by using Hyde's platform to build a specialist regulatory code-gen model trained on historical reports, analyst expertise, and 35TB+ of trading data.
Context
U.S. exchanges and equity market participants are required to produce reports for regulators like the SEC, detailed disclosures that analyze execution quality and order routing practices across massive volumes of trading activity.
For this U.S. Stock Exchange, generating these reports required analyzing more than 35 terabytes of execution and market data spread across dozens of internal and external systems. Producing the required reports relied heavily on manual workflows and static reporting (e.g. large PDF and CSV files), with human analysts spending weeks aggregating data, validating calculations, and ensuring compliance.
The exchange needed a system capable of interpreting massive trading datasets rapidly while replicating the reasoning of its top regulatory analysts.
Solution
The U.S. Stock Exchange used Hyde’s platform to build a specialist regulatory AI model capable of generating and validating SEC reports directly from raw trading data.
The first step was creating “golden datasets” based on historical reports and common analyses workflows developed by subject matter experts. This was enriched by the integration of more than 50 raw tables and APIs into a unified trading ontology using Hyde’s Codegen. These datasets captured the analytical reasoning and calculation patterns used by regulatory analysts when interpreting execution data.
To train the model at scale, the Hyde platform simulated 10,000+ reinforcement learning environments using its proprietary Trajectories tooling. This simulated analyst expertise which allowed the system to learn how to generate regulatory analyses under a wide range of market conditions. Outputs were evaluated using state-of-the-art model verifiers, which scored results against a rubric of human-defined evaluation criteria. Using the distilled outputs from these simulations, the platform fine-tuned a leading OSS model that encoded the reasoning pathways of regulatory analysts.
Performance continually improves through an ongoing reinforcement learning loop, where usage patterns and analyst feedback automatically become new training signals, allowing the system to adapt as market conditions and regulatory requirements evolve.
Impact
The system transformed how the U.S. Stock Exchange manages regulatory reporting and market analysis. The model now generates reports with complete accuracy, dramatically reducing operational risk while eliminating weeks of manual analysis work.
Improved visibility into execution performance has also strengthened commercial negotiations with market makers, generating incremental revenue through improved payment-for-order-flow economics. Beyond internal use, the platform has become a strategic asset, providing the opportunity for the U.S. Stock Exchange to offer the product to equity market participants and brokerages, representing an additional $20M+ potential revenue opportunity.
Takeaway
Only specialist AI models have the capability to deliver the superior accuracy required to meet the stringent regulatory requirements in the financial sector, due to their ability to reason across massive financial datasets.
The constrained size of a specialist model enables efficient data integration at scale and allows for millions of simulations to be processed quickly, ultimately producing precise outputs that continuously improve.
