Eventus Systems Uses $10.5m Funding to Advance AI, Double Workforce

Trade surveillance and market risk platform provider Eventus Systems is using a recent $10.5 million funding round to increase its use of artificial intelligence (AI) to deliver more concise alerts from large datasets, and to make more critical hires.

Eventus CEO Travis Schwab says the vendor has doubled in size and hired 11 people over the last two months, and plans to expand even more.

“I expect us to add at least another 10 people before the end of the year, maybe even a few more. … Other areas of the spend will go to data and connectivity,” Schwab says.

Recent critical hires include Perry Barth as CFO and Dan Burton as vice president of engineering. The firm is currently hiring for account executives, relationship managers, and engineering roles across the US, Europe and Asia-Pacific.

In the first quarter of this year, Eventus raised $10.5 million in a Series A investment round. Investors included a range of venture capital and financial services organizations, led by Chicago-based Jump Capital and Austin-based LiveOak Venture Partners. Also participating in the round were existing customers and investors, including Coinbase Ventures and Chimera Securities.

Eventus’s product, Validus, helps firms monitor all trading surveillance requirements and is available as a real-time or T+1 cloud-based or on-premise solution.

The company provides clients with a user interface or a dashboard, and the program can then be run through a web browser. Validus acts as the host site. Once the client is logged in to a tailored web portal, they can set up windows and applications that will run off the Validus browser.

Validus currently has hundreds of pre-built surveillance procedures to catch signs of spoofing, layering, wash trading and other types of market manipulation.

The software processes trade lifecycle messages in real time across an array of asset classes including equities, options, futures, fixed income and foreign exchange. It then uses machine learning and a procedural approach to reduce alert noise.

“We have narrowed our application of AI to alert resolution, and we use AI alongside a procedural approach,” Schwab says.

The AI sifts through a large dataset and then flags what it may consider to be spoofing. That alert is passed on to a human for investigation.

Schwab says systems can generate lots of noise if flagging parameters are set wide enough to catch what could be spoofing, so ” we spend a lot of time trying to generate what we call actionable alerts.”

These systems, not just ours specifically, can generate a lot of crappy noise and crappy data,” Schwab says.

Validus aims to fine-tune alerts and procedures to generate higher quality data that can result in actionable alerts. After that process, another layer of AI runs on top to go through the alerts and single out information the client should focus on.

With the new capital, Eventus plans to further enhance the actionable alerts generated from the system’s AI, put money into working with larger datasets and ultimately reduce noise for customers.

Schwab says market manipulation is a big focus for trade regulators, and that spoofing and layering are key techniques to manipulate the markets. He says using a combined machine-learning and procedural approach helps to explain these instances to regulators.

“If you can’t explain to a regulator why an alert was generated and what action you took to resolve that alert, you’re going to be in real trouble,” Schwab says.