Finance
FX Trading

Benchmark analytics extended across asset classes.
Calculations moved to AWS, costs cut.

Machine learning now predicts the most profitable FX benchmark for each risk portfolio, running on infrastructure migrated off physical servers and onto the cloud.

ML models predict
benchmark profitability
AWS replaces physical
server infrastructure
TCA extended to more
asset classes
Advisory on AI-driven trading
and risk strategy

Client

A global foreign exchange trading provider, running benchmark analytics and transaction cost analysis (TCA) across its FX pairs.

Goal

To enhance benchmark analytics for foreign exchange pairs and improve transaction cost analysis across additional asset classes.

One benchmark system, three things it couldn’t do.

The client’s benchmark analytics worked for FX pairs, but the business wanted to cover more asset classes without rebuilding the platform from scratch. That meant extending existing systems rather than replacing them.

On top of that, choosing the right benchmark for a given risk portfolio was still a manual judgement call. And the calculations underneath it all were running on physical servers, which capped how far the system could scale.

Three constraints were non-negotiable:

  • Extend existing systems to cover additional asset classes
  • Predict the most profitable benchmark for each risk portfolio
  • Migrate calculations to the cloud for scalability and efficiency

Cloud infrastructure and a model that picks the benchmark.

The calculations behind the benchmark analytics were migrated off physical servers and onto AWS, giving the platform room to scale without the operational overhead of managing hardware.

The benchmark stopped being a manual call once the model could tell you which one would actually pay off.

Alongside the migration, machine learning models were built to predict which benchmark would be most profitable for a given risk portfolio, turning a manual judgement call into a data-driven one.

On top of the build, the team provided high-level advisory on AI-driven trading strategies and risk management, using data science tools to sharpen the client’s decision-making beyond the immediate project.

Broader coverage, sharper predictions, lower costs.

Multi-asset Benchmark analytics extended to additional asset classes, improving the accuracy of profitability predictions for risk portfolios.
Cloud-native Calculations migrated from physical servers to AWS, significantly enhancing scalability.
Lower costs The AWS migration reduced operational costs tied to running the benchmark infrastructure.
Sharper strategy Data-driven insights optimised trading strategies and strengthened risk management practices.

The client now covers more asset classes than before, predicts benchmark profitability with a model instead of a guess, and runs it all on infrastructure built to scale.

* Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.

More case studies.

Your biggest pain point.
Fixed in 14 days. 50% off.

This started with one conversation. Book a 30 minute brainstorm call — we’ll plan your first AI project together and issue your 50% discount code. No payment today.

Book your 30 min brainstorm call → (free — we’ll plan your first AI project together)

*Case studies reflect work undertaken by our Heads of AI either during their tenure with Head of AI or in prior roles before they were part of the Head of AI network; they are provided for illustrative purposes only and are based on conversations with our Heads of AI.