Tech | Visa | Scholarship/School | Info Place

The risks and rewards of generative AI in software development

Join us in Atlanta on April 10 to explore the future of a safe workforce. We’ll explore the vision, benefits, and use cases of artificial intelligence for security teams. Request an invitation here.

As a 20-year coding veteran and the CEO of a company that provides services to software developers, I have an instinctive skepticism about early predictions that generative AI will eventually make most software development skills become obsolete.

While I’m still somewhat skeptical, my experience using gen AI in my day-to-day development work has prompted me to open my mind to what I think is possible. Artificial intelligence will change software development in some very fundamental ways, for better and for worse. Let’s start with the positives.

End the hard work

Developers spend a lot of time on details like syntax and punctuation. Much of this can (and should) disappear. Rather than poring over a manual or piecing together pieces from a code exchange, they describe the desired results and get perfectly formatted code in response. Large language models (LLMs) can also inspect existing code for typos, punctuation errors, and other details that drive developers crazy.


Software frameworks like Spring, Express.js, and Django greatly improve productivity by abstracting the mundane aspects of software development, setting consistent guidelines, and providing pre-written code for common functionality. Gen AI will enhance its value by creating boilerplate code, automating repetitive tasks and suggesting code optimizations. AI can also help customize framework components for specific projects.

VB event

Artificial Intelligence Impact Tour – Atlanta

Continuing our tour, we will head to Atlanta for the AI ​​Impact Tour stop on April 10th. This exclusive, invitation-only event in partnership with Microsoft will discuss how generative AI is transforming the security workforce. Space is limited, please request an invitation now.

request an invitation

The rise of the generalist

For many developers, their advantage lies in their expertise in a specific language. When machines can spit out code in any language, proficiency in Python or Ruby becomes less important. Likewise, professional back-end skills such as testing and code optimization will migrate quickly to new generations of AI models. The most valuable skills will be those that machines are not good at, such as building compelling user interfaces, translating user needs into specifications, and inventing new ways to support customers. Software “poets,” or people who dream up great ideas that can be realized with code, will be in the spotlight.

testing revolution

Gen AI is designed for software testing. Developers write the code and the bot creates as many test scripts as needed. A recent IDC survey found that the two most anticipated benefits of next-generation artificial intelligence are software quality assurance and security testing. This will disrupt DevOps practices of continuous integration/deployment and push many testing experts to find new areas of work.

Civic development is in full swing

Current low-code/no-code development tools are already great, gen AI will take them to the next level. Despite the elegance of automation, low-code/no-code still requires people to piece together workflows on a whiteboard before committing them to software. In the future, they will be able to provide a hand-drawn sketch of the desired workflow for their model and get the necessary code in seconds.

But artificial intelligence is not a panacea

Despite its many promises, this new generation of artificial intelligence should not be viewed as a panacea. Consider these potential disadvantages.

The risk of over-testing

Since the model can generate tests quickly, we may end up with more results than we need. Over-testing is a common problem in software development, especially in organizations where performance is measured by the number of tests produced by the team. Running too many duplicate or unnecessary tests can slow down a project and further bottleneck the pipeline. When AI can suggest when to remove tests, we’ll see a flood of developers unblocked — the vision of a generation of AI that makes me excited for the future.

Skill degradation

“I will always choose a lazy person to do a hard job because he will find an easy way to do it.” This quote is often incorrectly attributed to Bill Gates. While the origin of this quote is unclear, the sentiment makes sense. Lazy people look for shortcuts to avoid hard work. Gen AI is an antidote to lazy developers. It can lead to the creation of bloated, inefficient, and poorly performing code. It stifles the innovation that makes good developers so valuable. Remember, a generation of AI is coded based on existing patterns and data. This may limit the innovation potential of developers, who may not consider more out-of-the-box solutions.

trust deficit

Gen AI is only as good as the data used to train the model. Poor quality data, training shortcuts, and poor hint engineering can result in AI-generated code that doesn’t meet quality standards, is buggy, or doesn’t do the job. This could lead to organizations losing trust in the quality of this new generation of AI and missing out on its many benefits.

Now comes the money question: Will artificial intelligence make software developers obsolete?

Although some high-profile experts have advanced this view, there is no historical precedent for such a conclusion. Technological advances—from high-level languages ​​to object-oriented to frameworks—have steadily made developers more productive, but demand has only increased. Gen AI may undercut the market for low-end basic coding skills, but the larger impact will be to move the entire profession up the value chain to do something that LL.M.s are currently less good at: innovation. Remember, new generations of AI models are trained on what is known, not what is possible. I don’t expect machines to design revolutionary user interfaces or dream up Uber anytime soon.

However, developers won’t see such a transition again in their careers. Instead of getting mad at the machine like I initially did, they should just roll with the punches. The prospect of eliminating much of the tedium of building software should excite everyone. The risk that some features may disappear should be an incentive to take action. Highly qualified developers who transform business requirements into elegant and performant software will always be in high demand. Make it your mission to improve your skills.

Keith Pitt is the founder and CEO of Buildkite.

data decision makers

Welcome to the VentureBeat community!

DataDecisionMakers is a place where experts, including technologists working in data, can share data-related insights and innovations.

If you want to stay up to date on cutting-edge thinking and the latest information, best practices and the future of data and data technologies, join us at DataDecisionMakers.

You might even consider contributing your own article!

Read more from DataDecisionMakers

#risks #rewards #generative #software #development

Leave a Reply

Your email address will not be published. Required fields are marked *