DeepNext will support developers with autonomous application development in financial services.
STX Next, a global leader in IT consulting, has announced that its virtual development solution, DeepNext, has entered beta testing at Newable, a UK-based financial services organisation.
DeepNext will be applied directly to real-world problems and support Newable in reducing technical debt, by handling repetitive tasks that can be performed quickly by the tool. DeepNext will add capacity and ensure human developers can work on more high-priority and intensive tasks that require increased cognitive input.
First launched in March 2024, DeepNext was designed to assist businesses with the maintenance and growth of their applications. The solution acts as an autonomous AI developer, project manager or QA engineer, handling tasks with high levels of complexity all the way to deployment.
Now, the tool has transitioned from laboratory testing to beta testing in a real-world setting in the financial services industry. A team of researchers will also join the project to monitor the system’s development and ensure testing is extensive.
Michał Skibicki, AI Manager at STX Next said: “The ability to test a solution such as this in real-world business scenarios is crucial for its improvement. Testing is an integral part of the process and will assist in ensuring that users, developers and managers can truly rely on DeepNext to handle autonomous tasks and become a valued part of any team.”
Bartek Roszak, Head of AI at STX Next said: “In multi-agent solutions like DeepNext, it is essential to design the architecture optimally. Most multi-agent solutions are monolithic, where each component depends on the previous one and cannot be isolated. DeepNext deals with the code development process which can consist of many stages, and we can divide each stage into separate modules.
“With this approach, we are able to improve and test individual system components without needing to run the entire application. Such a system is easier to monitor and diagnose, making it simpler to identify at which stage the system failed in case of an error.
“In beta testing, the key advantage is that we can diagnose which elements of the system need adjustment for each real-world case. Thanks to the implementation of a modular approach, we can quickly make adjustments to improve the system.”
For more information go to: https://www.deepnext.ai/
