Imposter syndrome—doubting your abilities to the point that you feel like a fraud—is an evergreen topic of conversation among software developers. For many devs, the explosion of GenAI and AI-powered coding tools makes feeling like an imposter more inevitable than ever. Plenty of people who code for a living are scrambling to add AI prompt engineering and other related skills to their repertoires and worrying about their essential job functions being taken over by AI.
In this moment, how can tech leaders empower their developers? By building a culture that recognizes and rewards continual learning and treats GenAI as a powerful addition to the developer toolbox: a way to automate toil and expedite the acquisition of new skills, allowing developers to work more capably and creatively.
Why so many developers feel like frauds at work
No industry is immune to imposter syndrome, but certain aspects of how software developers work can leave them particularly vulnerable to feelings of imposterism.
There’s always something to learn
Technology and best practices are constantly evolving, which means that software developers have to remain open to acquiring new skills or polishing existing ones, rather than believing they have nothing left to learn. There’s alwayssomething new to learn—which means there’s always something you don’t know how to do.
It can be hard to learn incrementally
People tend to gain more confidence in their abilities when they can acquire new skill sets incrementally, according to Dr. Cat Hicks, Director of Pluralsight Flow’s Developer Success Lab. However, software engineering doesn’t always seem to reward or even allow incremental learning. “It’s just ‘learn Python,’ or ‘learn React,’” Hicks says.
The industry can be a pressure cooker
Many devs also feel intense pressure to re/upskill with whatever time is left over from their day jobs. They spend their time away from work learning new languages, contributing to open-source projects, and compiling a portfolio—working, in other words. For plenty of developers, it feels like the choice is between sacrificing necessary recharge time and non-work obligations or faltering in their careers.
Especially with tech influencers sharing their side hustles or hobby projects on social media, some devs start to feel like everybody is working on something more complex, creative, or innovative than they are. And with so many developers working remotely, it can be hard for developers to form a realistic picture of how their peers are working or how they’re really spending their time.
Learning is part of the job
The runaway pressure on devs to learn new skills, languages, and frameworks can trap them in what Hicks describes as a stress cycle, “a form of physiological conditioning where you associate learning with high-stress environments.” When learning seems stressful, high-cost, and low-reward, people avoid situations where they’re challenged to develop new skills: a vicious cycle that amplifies feelings of imposterism.
Stack Overflow’s annual Developer Survey has shown that access to learning opportunities at work is very important to devs. But while many organizations pay lip service to developers’ desire to learn at work, they too often disincentivize learning by focusing only on devs’ quantifiable output: codes, commits, and PRs.
In a qualitative research project involving more than two dozen software developers and engineers, Hicks found that “code review often did not recognize code writers’ effort when it did not result in lines of code.” In spite of “stated ideals about knowledge sharing,” Hicks writes, “this work was often contradicted with negative cues from colleagues about what was ‘truly’ valued.” As a result, code writers felt lonely and adrift, a state one developer described as “like coding in the dark.”
The genAI era
The rise of AI coding tools has the potential to exacerbate this disparity. Instead of creating the time and producing the resources required for their developers to learn at work, some employers are going to try bridging skill gaps with AI coding tools, which in a sense offer nothing but quantifiable output. GenAI doesn’t suffer from imposter syndrome, though perhaps it should. It’s been clearly demonstrated that large language models (LLMs) trained on incomplete, inaccurate, or out-of-date information are likely to produce erroneous and misleading results. Knowledge managementis as important for AI as it is for humans.
Rather than seeing the implementation of AI tools as a replacement for human learning, tech leaders who want to empower their developers should think of AI as a powerful tool for accelerating learning and upskilling while automating toil.
To learn more, visit us here.