Given this scenario, automation is presented as the unified solution the problem requires. Features such as semantics and process automation can offer their benefits to the changes that will be performed in different areas: business services, online services and data management.
The FinTech phenomenon involves changes in many different areas with many different technologies. This leads to a complex scenario where the problem must be addressed from a global perspective, otherwise there is a risk of running into an operational and technological chaos.
New techniques in Software Automation suggest this methodology as a good solution to tackle the challenge.
Software automation raises the technical solution to the conceptual level where operates the business. This higher point of view provides the more comprehensive and unified approach required to address FinTech changes.
Working in a more abstract level allows not only the automation of technical process but also the automation of traditional business processes. Decision engines, risk based pricing, default prediction and fraud detection are examples of processes that can be improved under this perspective.
The new way of defining programming languages, based on the concepts that structure them, opens the door to Knowledge Representation. This discipline, a traditional field of artificial intelligence, is increasingly being applied to data processing (visualization, searching, analysis, decision making …) and in the future will be essential to any process of analysis and interaction, either with customers or with other systems. XBRL– eXtensible Business Reporting Language – and FIBO – Financial Industry Business Ontology – are examples of semantics applied to financial industry
3.Less technical complexity.
Again, the higher level of abstraction of the languages allows to solve, automatically, much of the technical details inherent to the different solutions. This automation of complexity results in quality, safety, standardization and other software features.
Having the control of programming languages, the same development can be translated into multiple platforms without having to perform a new project for each new device.
Once the programming languages are closer to the business, developers can focus on providing a solution from the business point of view and not from a technical perspective. It also breaks down the traditional communication barriers between the development areas and the rest of areas of the company.
FinTech is working on multiple areas in which the automation can provide advantages. We can group them into three main blocks: business services, online services and data management.
Meeting customers’ new demands and productivity improvement are objectives that are moving the industry to act over their more traditional structures. Automation provides advantages on:
Portfolio management. The automation of processes provides a more streamlined management. On the other hand, the semantics allows a unified management of heterogeneous portfolios and the integration of external ones.
Compliance. Semantics, along with the use of business-oriented languages facilitate audit analysis processes and the deployment of its documents. Beside this, process automation is the best way to solve compliance issues.
Legacy systems. The higher level of abstraction provides a straightforward way to design interfaces with legacy systems.
The emergence of new devices, the improvement of communications and a new digital culture have made institutions open their doors in order to supply a more interactive management to their customers. Again automation provides benefits in this area:
Financial education. Semantics gives the systems documentation and visualization capabilities facilitating its handling and understanding.
Personal Management. As a result of the previous points, the customer can perform a more personal management of their data and assets. As mentioned, the semantics play a major role in the interaction with the customer.
New Banking. Multi-platform, semantics and process automation enable new business lines such as personalized offers, payment platforms, crypto-currencies, proactive risk analysis and others.
A more interactive relation with the customer and the new methodologies to process that information have revolutionized traditional CRM techniques and the whole data management. Automation benefits next areas:
Reporting. Semantics is increasingly being a part of reporting. XBRL is a good example.
Security. Process automation enhances the security in data processing.
Big Data. We can summarize – this point would require an entire article – that semantics, process automation and a higher level of abstraction allows, on one hand, to handle the variability of the data and, on the other, mitigate the risks associated with the unstructured schemas of Big Data.
The financial sector is facing one of the greatest technological challenges in its history, involving changes in many different areas with many different technologies.
Software automation provides the global and unified approach required in this scenario, otherwise, individual approaches can lead the companies to a technological landscape very hard to manage.
We will use the Language Workbench, once the languages (DSLs) are defined, to perform the programming process, and along this process we will generate code to compile it and check the result in the final application.
Whenever we use languages that support partial classes, abstract classes, generic types, delegates, etc., we can generate code safely because we will do it in separate files without risk of overwriting other code such as manual or FW code. In other kind of languages we will need to use other strategies to mitigate this risk.
Merging with Frameworks
As in the previous case, depending on the final language in which we generate, this process will be more or less simple. In the case of FW-friendly languages (abstract and partial classes, delegates, etc.), this merging process it is provided by the syntax itself.
In other kind of languages we will need to adapt some functionalities or some configuration files, which can also be automated by the LW.
In most cases there will be certain features that may not be automated; otherwise our DSLs will be too complex, losing, as a result, their power.
In order to integrate the whole process on the LW, the best scenery is defining these extension languages on it, but, If this is not the case, we will program these extensions in the final language where we generate the code.
If we program the extensions in the final language, the most suitable way is to give the generated code a FW structure. With this extensible approach we avoid the risk of loss or replacement of the code.
Once we have all the code: FWs, generated, manual extensions, configuration files and other, an automatic process performs the merging -if necessary-, the compilation and the deployment.
Depending on the operating system will have different tools to do it, but normally we will use batch files to program all the steps of this process.
In order to conclude, we want to remark that we have explained the methodology the way is done nowadays, but it is worth noting that it is a constantly evolving technology that aims to integrate the whole process in the LW: languages, debug, merging, building, etc.
View all Academic Posts.