Taking a brief virtual walk in a Cybersecurity Architects' shoes this past week allowed me to exercise an understanding of what it may take to secure and evaluate a system for a financial institution. As a developer, architect, and technical advisor I have implemented payment systems like WooCommerce, coded monetization APIs with Stripe, and fortified Payment Card Industry Data Security Standard (PCI DSS) compliance for Enterprise Resource Planning (ERP) systems.

Knowing how all these applications lend themselves to modern Software as a Service (SaaS) products like Shopify, changes the range of liability to the business, and simplifies the demand of regulatory practice that was initially applied from eCommerce. When software becomes a product and payments become a platform, the responsibility shifts, replaced by cost, but a lot less direct liability.

Financial systems from online payments to banking are required to meet an extensive list of compounding regulations that have carried on from the brick and mortar facilities they once shaped, to the digital systems that increase their reach. From the browser to the database, Financial Technology (FinTech) institutions are known to carry their litany of requirements which has forced many out, and reshaped the few that remain.

Above all of the legal regard and overhead, sits a layered system of protective measures that supports the business in rising to the mandated requirements of a financial operation. As these systems scale and enumerate a planned set of practices is required to safe guard the business and its customers from adversaries known and unknown.

As a principal of enabling a cybersecurity focused platform, a layered approach begins with the physical or structural considerations that often deter and mitigate risk in a direct manner, with fencing, key-fob entry, and monitored surveillance by security staff. Next in line is protecting all the buildings and data centers with private security and physical safeguards to establish a base layer of detection.

At the perimeter of the infrastructure, a business needs to established network and application firewalls in addition to packet filters, designed to block and log inbound threats to private networks. On the internal network endpoint security and firewalls, block and detect intrusion that has occurred from inbound, and on-premises attacks.

When moving towards the code, application security is enabled through a series of secure by default practices including, Common Vulnerabilities and Exposures (CVE) build scanning, secrets management, Cross-Origin Resource Sharing (CORS) configurations, Cross-site Request Forgery (CSRF/XSRF) protection, enforced Transport Layer Security (TLS) traffic, and Distributed Denial-of-Service (DDoS) web server protection.

Protecting the data security and integrity begins at the exchange through the application, with in transit encryption over private subnets, and then in storage and access, encrypted at rest, with automatically rotating credentials for accessing the databases. An imperative step is for the business to implement policies and procedures that are delivered through quarterly trainings, aligned with Privilege Access Management (PAM) practices and are enforced by Identity Access Management (IAM) in the cloud Identity Provider (IdP).

Focusing on the ability to secure and implement more automation in the infrastructure to detect misuse by customers and potential threats in the system is a common goal of all businesses, when this is applied to FinTech, the stakes are even higher. As Artificial Intelligence (AI) begins to specialize in different tasks, it becomes increasingly important to consider how AI agents can be used to analyze bad actors and intrusion in real-time. Cloud providers are beginning to offer an increasing amount of Agentic AI utilities that can gather and return information for analysis, which can process, and remediate in a Zero Trust model much quicker than human counterparts.

The augmentation of these systems will allow for businesses to quickly mitigate and address an attack, based on patternicity and evaluation that is being continuously processed from an Autoregressive (AR) and Adversarial AI model. As vulnerability databases gain insights on new vulnerabilities and threats from AI Agents, risks can be increasingly tracked through the Single-Pane of Glass (SPOG) dashboard in alerts, shared across distributed networks, tagged, and awaiting the approval for auto remediation or escalation. As the alerts are processed cyber response teams can prioritize them through risk rankings and utilize mitigation with AI agents.

As a Financial Technology company that provides services in loans and savings, the business may be required to proceed with the upmost diligence in Know Your Customer (KYC) and Know Your Business (KYB) processes. The financial firm must confirm that their customers are who they say, monitor their actions, and identify how the money being processed is being used to validate its authenticity.

When approaching the many laws and regulatory practices of Financial Technology, business must see to the most stringent security for protecting consumer, financial, and monitoring data that is stored in the system. To properly ascertain some of the above methodologies FinTech companies must create algorithmic detection within software applications, secure that data, and encrypt user interactions to protect that data. Cloud systems and access to this critical infrastructure must be continuously audited to enable a clear distinction of the roles from PAM and their intentions for use through approval with IAM.

As a financial businesses presence in cyberspace continues to transfer from the often brick and mortar setting, it is paramount that their actions match that of the considerations in the digital space. They must be ready to handle attacks before they occur, anticipating the levers that will allow for mitigation of the risk and threats, while anticipating the vulnerabilities in a 360-degree point of view.

AI can move faster than humans, the work that the computer processes, needs an expert to adapt to the situation, this is a shared responsibility between the human and machine interaction. When supporting such a vast lake of data, it is important to continuously scale the security and monitoring efforts in the digital space, enabling FinTech to be one step ahead of advisories, and more competitively reliable for its' customers. Integrating instead of only manually reviewing, is only the beginning of benefits to implementing active and passive countermeasures enhanced by AI driven security.

All advice is temporal: subject to change, open for review, often only slightly — Follow along on X, GitHub, Instagram, LinkedIn, and YouTube.