Software Development in the Age of AI

Much has been written in recent years about how AI is accelerating software development. And it’s true. Tools like Claude, GitHub, Copilot and other AI-driven environments reduce the time needed to write code, set up test scenarios, and handle repetitive tasks. But for us at Dolphiq, that’s not the most interesting part of the story.

We’d rather ask a different question: what will you do with the time AI gives back to you?

Artificial Intelligence

Author: Allard Rothengatter

Published: 20 March, 2026

How We’ve Been Using AI in Software Development Since 2022

At Dolphiq, we’ve been actively working with AI in software development since late 2022. While most were using AI for minor optimizations, we were already applying it to review existing code, improve isolated pieces of JavaScript, and analyze logical structures. Always with the same guiding question: is this production-ready?

With over 26 years of experience in software development, we had our own benchmark to measure that output. Not to dismiss AI, but to understand its limits and how quickly they were shifting. And they are shifting fast. Where AI was once only useful for code snippets of 30 to 40 lines, we now see it generating entire features and first product versions that serve as a foundation for further development.

From Prototype to Reliable Product: An AI Chatbot

An AI-driven chatbot as a real-world example

A concrete example of developing software with AI is a project where we built an AI-powered chatbot for searching product data. Chatbots have been around for years, but traditionally they’re limited to static answers to predefined inputs.

In this project, we used AI to automatically enrich product data with metadata, combine information from multiple sources, and generate answers that directly match the user’s question. Instead of returning a list of search results, the system analyzes relevant sources and provides a single concrete answer.

How did AI help with this chatbot, and what did we do ourselves?

What AI did in this project: set up an initial prototype, propose data structures, and generate an initial chat interface. What we did: define the framework, safeguard the architecture, validate the output, and tailor the solution to the client’s specific business case.

AI built the first version. We turned it into a product that will still work two years from now.

AI enhances our work, but it doesn’t replace our creativity and expertise.

– Allard Rothengatter, CEO Dolphiq

Why AI Software Development Takes Testing to the Next Level

One of the areas where AI-driven software development has made the biggest impact recently is testing. Every new feature increases the complexity of an application, and in systems with hundreds of thousands of records, the risk of unexpected regression is significant.

We now use AI to generate test scenarios, set up test coverage, and analyze edge cases that might otherwise be overlooked. The direct result is more test coverage with less manual work. But the indirect benefit may be even more valuable: it gives us the space to think about what we’re testing and why, instead of just executing tests.

Making Complex Government Data Accessible to Citizens

The challenge of the Open Government Act

A second project illustrates how the collaboration between AI-driven software development and human insight works for truly complex problems. Under the Open Government Act, large volumes of government documents must be made public: emails, decision documents, internal communications. Tens of thousands to hundreds of thousands of files per project.

The technical challenge is already significant: making all that data searchable. But the real challenge is ensuring the information is understandable to citizens without a bureaucratic background.

AI structures, we design the experience

AI analyzes and structures unstructured data, enriches documents with metadata, and draws connections between sources. We design the search experience, decide how information is presented, and ensure the context is correct. The result isn’t just a search engine, but a translation layer between government data and the citizens who use it.

AI Output Is Not the Same as Reliable Software

At Dolphiq, every code change goes through a structured process of pull requests, peer reviews, and validation. AI output is no exception. In fact, we often scrutinize AI-generated code even more closely, precisely because it can look solid at first glance.

Good software isn’t just about what works now. It’s about what will keep working as the application grows, as the organization changes, as new integrations are added. That kind of durability isn’t built by generating code quickly. It’s built by understanding the foundation.

What Are the Limitations of AI?

What AI Can Do

Write, review, and test code. Analyze and structure data. Quickly generate prototypes and solutions. In short: the execution work.

What AI Can’t Do—But We Can

We look beyond the execution. What does the client truly need? Answering that requires depth, expertise, and the willingness to have the ‘tough conversation’ when needed.

The Biggest Advantage of AI? More Time and Space to Do Better Work!

The biggest change AI-driven software development brings to Dolphiq isn’t about speed. It’s about the nature of the work. Developers who used to spend much of their time on execution can now focus more on architectural decisions, advising clients, and critically thinking through a solution before writing the first line of code.

That’s the real gain. Not the seconds an AI tool saves when generating a function, but the space it creates to do better work.

Conclusion: AI Software Development as a Tool, Not a Replacement

AI software development is changing the way we build software, but not in the way many people expect. It’s not a replacement for the developer, but a tool that frees up time for the work that truly matters.

At Dolphiq, we use AI-driven software development to prototype faster, test better, and spend more time on the questions behind the technical assignment. Because the greatest value isn’t in the code AI generates. It’s in the quality of the decisions you make when you’re no longer spending all your time on execution.

Curious about what AI can do for your organization? We’re happy to help.

Naam

Bedrijfsnaam

Telefoonnummer

Bericht

How do we collect and treat your data? Read all about in our Privacy statement

More interesting blogs

ChatGPT-4: What We Know and What We Can Expect

28 Feb 2023 by Alan Linnenbank

The release of OpenAI's GPT-3 language model in 2020 marked a major milestone in the development of artificial intelligence. With its impressive 175 billion parameters, GPT-3 demonstrated unprecedented capabilities in natural language processing and text generation. But what comes next? Will OpenAI continue to push the boundaries of AI language models with ChatGPT-4, and if so, what can we expect?

The Power of ChatGPT

25 Jan 2023 by Allard Rothengatter

ChatGPT is a smart chatbot developed by OpenAI. The model is trained on millions of texts from the internet and is able to understand and generate language in a way that was previously impossible. In this article, we will discuss some of the main benefits of ChatGPT and how it can change the way we communicate and process information.

ChatGPT o1: The new standard in AI language models?

17 Sep 2024 by Allard Rothengatter

In the dynamic world of artificial intelligence, one name is synonymous with innovation: OpenAI. With the launch of ChatGPT o1, also known as the “Strawberry Model,” they are taking the next step forward in AI-driven communication. This advanced language model not only promises a leap in efficiency and flexibility but also introduces various additional capabilities that will inspire both clients and developers.

ChatGPT o1 is designed to seamlessly align with the needs of modern organizations. Whether it’s improving customer interactions, accelerating development processes, or discovering new forms of creativity, this model offers a range of possibilities that raise the bar for what AI can achieve.