‘Still Very Much in Control’: Bernd Alter on the Current Impact of AI on Development
With the rapid and widespread integration of AI into our daily lives, it is essential to give due consideration to its impact on developers. Over and over again, we see the growing sentiment that AI is displacing developers from their roles. “It definitely changes the way we work, but the role of developer will stay in the future too,” says Bernd Alter, Co-CTO at Turbine Kreuzberg. This convergence of technology and expertise not only reshapes the way businesses operate but also challenges developers to push the boundaries of their skills and learn more. To get a read on the current situation, we asked Bernd to answer our questions.
Bernd, how has the daily routine of the developer changed so far?
AI’s influence on a developer’s daily tasks is pretty significant, although it hasn’t profoundly altered the core foundations and everything yet. What’s interesting is the transformation in the research aspect. AI has sped up tasks like looking up how to work with data structures. For example, googling quick-forgettable things like “how to flatten a multidimensional array” or “how to extract a certain thing from an array” can sometimes be time consuming. AI tools like Github’s Co-Pilot and Amazon’s Code Whisperer not only provide quick answers, but also anticipate a developer’s needs, suggesting code and structure.
It’s important to note that developers are still very much in control, which proves that AI is in the role of assistance, not a substitute for developers. In essence, AI is streamlining and enhancing the development process, making developers more efficient.
In your view, how do you see the role of a software developer changing? What new skills should we focus on now?
I certainly see the landscape of software development having evolved — and it’s quite fascinating. Even if you’re not well-versed in a specific programming language, today’s AI can provide you with precise code suggestions, which is a game-changer.
As for what developers should focus on in this evolving environment, skill advancement is key. AI is becoming proficient at handling routine coding tasks, sometimes matching the quality of human work. This raises the question of whether AI might replace certain developer roles. However, I believe this will never affect developers who have a strong foundation especially in software design, architecture, and patterns. These skills go beyond coding and will set you apart from AI, because they also require creativity in applying them, something AI is not capable of (yet). So, it’s an exciting time to prioritize these skills to maintain a competitive edge in the field.
Do you have tips for communicating with the AI/prompting? How can you best integrate the AI tools in your daily tasks?
Developers now have the advantage of seamlessly integrating AI tools and assistants into their Integrated Development Environments (IDEs). It’s all about ease and accessibility. In many cases, you simply need to enable these tools and use them within your workflow. They often operate by merely interpreting a command or description of what you intend to code, and then they generate the code for you. It’s a matter of harnessing these capabilities and seamlessly incorporating them into your development process.
How can developers use AI power to ease their complex tasks?
You know, AI is a pretty remarkable tool for developers. It’s not so much about tackling complex problems but rather taking care of the repetitive and straightforward tasks. AI plays a crucial role in freeing up developers and allowing them to focus on more complex challenges. It’s a valuable tool when entering new territories within development.
However, it’s important to remember that AI isn’t a creative force; it works with existing knowledge and solutions, so it can’t invent entirely new solutions for novel problems. The role of a developer demands creativity, and human creativity will continue to remain indispensable. To put it into perspective: consider navigation systems like Google Maps. While they are usually accurate, there are also times when your personal knowledge leads to better and faster ways if you know the way.
So, instead of blindly following the recommendations without questioning, let the navigation guide you but try to learn the path to go further or faster on the road. Similarly, it’s essential for developers to learn and validate AI suggestions to ensure they align with their project’s specific needs and use them in further projects.
Where do you see AI and development headed in five years ?
It’s really interesting to see all the excitement surrounding AI these days. It’s like this new frontier of technology that holds so much promise, but there’s a catch — a lot of folks don’t fully grasp how it all works under the hood. It’s a bit like a well-kept secret held by a handful of experts. And that secrecy raises some concerns. If this trend persists, there’s a real risk that AI could turn into a playground for just a few big players, like Google and the rest, who hoard all the knowledge and dominate the scene.
When that happens, competition tends to fade away, and that’s where things could get tricky. Without healthy competition, innovation slows down, and the power becomes concentrated in just a few hands. To keep things fair and foster innovation, some argue that AI should be more accessible for everyone, not just a privileged few. It’s a real challenge, but finding that balance is key to the future of AI development.
Bernd, thanks for the chat!
