AI start-up Imandra has seen its Imandra Markets product – which creates ‘digital twins’ of exchanges to allow for testing – gain traction among European exchanges, driven largely by regulatory and commercial concerns. Paul Brennan, chief strategy officer at Imandra, spoke to BEST EXECUTION about how the firm tests these digital twins, what market pressures have led to its development, and where the technology could be used in other industries.
Imandra Markets digitises venue design, creating ‘digital twins’ powered by automated reasoning AI to ensure compliance and correct implementation, allowing exchanges and trading venues to test and improve system operational resiliency and meet regulatory obligations. Currently, Imandra Markets supervises exchange technology responsible for approximately 25% of all European equities trading, the firm said.
The digital twin is an approach used in other industries such as construction, manufacturing and aviation, where complex systems and physical assets are accurately modelled before being built and systematically monitored thereafter, helping to drive efficiencies.
A digital twin – in the context of complex systems – is a fully functional and logically precise virtual replica of any financial software system, which accurately simulates its operation and verifies its behaviour in real-world scenarios. “When combined with automated reasoning AI, a digital twin approach can increase operational resiliency while unlocking better business intelligence and innovation within financial markets,” Brennan told BEST EXECUTION.
“Breakthroughs in AI and mathematics allow us to model exchange rules and regulations precisely and apply rigorous logical AI to automate analysis and testing, all while providing logical audit trails.”
“Digital twins that can be interrogated allow exchanges to verify properties about the exchange system design, generate test cases to ensure the correctness of the exchange system, run an audit against production trading data and generate precise English-prose documentation,” Brennan said.
During the process of building the digital twin two key tasks are performed to make sure that stakeholder understanding is synchronised with the exchange design.
“We formally verify properties of the model. For example, if the exchange ranks orders on a price time basis, we will verify that the model respects this property always, via rigorous automated mathematical proof. We typically do this for all of the behaviours described in the exchange rulebook or trader manuals,” Brennan said.
“We use the model to generate test data to test the real exchange system in its own test environments. Through the testing process, we can pick up any discrepancies, which are investigated with the client exchange. Following these steps we can produce an accurate digital twin which models the actual exchange’s behaviour and identifies any discrepancies vs the design intentions. Once set up, this process runs on a daily basis to help ensure ongoing compliance.”
As to what regulatory and commercial pressures have led to this solution, Brennan said financial systems have become too complex for the traditional tools and methods used to design and test them.
“Exchanges and trading venue operators have a regulatory obligation to maintain the orderly function of the market, with system resiliency being a fundamental KPI. As we have observed, outages and failed technology upgrades are wide-reaching, with regulatory and commercial consequences, which can lead to reputational impact,” Brennan said
“The Imandra Markets digital twins offer significant advances in exchange system resiliency, accelerated system development, and open up avenues for business intelligence and new revenue streams.”
Brennan believes the technology has promise elsewhere, too.
“Our biggest footprint within financial markets is currently with exchanges and trading venues, however, there are wider applications of our technology. In fact, any newly designed, updated, or re-platformed financial system can come to market more quickly and more cost-effectively, if the system design is verified by Imandra. What’s more, the areas of operations that Imandra has verified enjoy a leap forward in governance.”
“Our automated reasoning AI has been successfully used in non-financial, safety-critical industries including US government departments. So our technology has a proven track record in other industries as well as the financial sector,” Brennan added.
©Markets Media Europe 2024
