- The Nevron Weekly
- Posts
- Handling Post-Launch Support, The Risks of AI-Built Software and Dutch Software Policy
Handling Post-Launch Support, The Risks of AI-Built Software and Dutch Software Policy
Nevron io
March 20, 2025
Welcome to the tenth edition of our weekly newsletter.
We would like to thank every single one of you who read last week’s edition, gave us feedback and we encourage you to keep doing so.
Have any questions you want us to cover? Reply to this email and we will answer them in depth next week!
This week we’ll talk about the following:
The Risks Of Building Software Without Technical Knowledge
Dutch Parliament Votes to Reduce US Tech Dependence
How We Handle Post-Launch Support and Maintenance
How AI Watermarking Makes Generative Content Traceable and Secure
The Risks Of Building Software Without Technical Knowledge
Tweet of Founder Sharing His Concerns
A founder who built a SaaS product with "zero hand written code" using Cursor initially boasted about its success—until security issues emerged just two days later.
Without technical knowledge, several critical vulnerabilities emerged:
- API keys maxed out: API (Application Programming Interface) keys control access to services. Without proper security measures, attackers exploited these keys, potentially causing service disruptions and unexpected costs.
- Subscription bypassing: Users found ways around payment requirements, directly impacting revenue because the developer lacked knowledge to implement proper authentication and payment validation.
- Database manipulation: People were "creating random shit on db" (database)—meaning unauthorized users could insert, modify or delete data, potentially compromising user information or corrupting the entire system.
The founder's admission that "I'm not technical so this is taking me longer than usual to figure out" highlights the fundamental problem.
While AI can generate functional code, it can't replace understanding the security implications of that code or how to properly configure authentication, data validation, and access controls.
Dutch Parliament Votes to Reduce US Tech Dependence
The Netherlands’ parliament recently approved several motions calling for reduced dependence on US software companies, including plans to create a cloud services platform under Dutch control.
Why?
Dutch lawmakers are getting increasingly worried about who has access to their data, especially with recent political changes in the US.
One representative from the pro-European Volt party put it bluntly, questioning whether Europeans are comfortable with "people like Trump, Zuckerberg, and Musk ruling over our data."
This is a concern many European businesses share too.
You Might Soon See
|
|
|
|
Recommended Precautions
Take time to assess your current dependence on US-based cloud and software providers while exploring European alternatives for your critical infrastructure and services.
Review where your data is stored and how easily it could be relocated if needed, and consider how your technology stack aligns with emerging European preferences.
Need help navigating these changes?
We can help you assess your current technology stack and develop strategies to adapt to evolving requirements. Reach out if you need support to transition.
Weekly Q&A
How We Handle Post-Launch Support and Maintenance
You asked and we’re happy to answer.
Post-launch support is a critical part of ensuring the success of any software project. Here’s how we approach it:
For business-critical projects, we work with clients to set up a fixed monthly fee for ongoing support. This covers agreed-upon services, while additional tasks outside the scope are quoted separately. Once the client approves the quote, we execute the work. | For non-critical projects, support is handled on a case-by-case basis. If further development is needed, we provide a quote for the work, and bugs are fixed free of charge. Security updates and software upgrades related to vulnerabilities are also covered at no cost. |
We value transparency—clients are never charged for anything without their explicit approval.
For emergencies like hacks or cyber leaks (not caused by us), we offer emergency support at a rate of €1500 per day to resolve issues quickly and minimize disruption.
Our Payment Terms
We keep payment terms simple and fair:
For one-off projects, we charge 50% upfront and 50% upon delivery.
For large projects (lasting over three months), payments are split into milestones—either monthly or based on milestone completion.
If you have questions about how this works or need tailored support for your project, feel free to reach out!
How AI Watermarking Makes Generative Content Traceable and Secure
AI watermarking embeds invisible cryptographic markers directly into AI-generated content, such as text or images, making it traceable and resistant to tampering.
Unlike metadata (external data tags), these markers are built into the content itself, ensuring they remain intact even if altered.
Illustration of the process of training an encoder-decoder watermarking model
This technology helps businesses verify the origin of AI-generated material, protect intellectual property, and prevent misuse or misinformation.
By using advanced techniques like error-correcting codes and pseudorandom algorithms, watermarking ensures security and accountability in generative AI.
Thinking About Tackling a Big Project or Adopting New Tools?
We’d love to help you plan and make it happen. Let’s talk about what’s next: Reach Out to Us.
Thanks for reading! If you have ideas or topics you’d like us to cover, drop us a message.
Stay in touch: Website | LinkedIn | Instagram
Don’t want to hear from us? You can unsubscribe anytime below.