Image by Anton Savinov on Unsplash
As someone passionate about computer science, I often hear the usual comments from people at school and on the internet, such as, “Dude, AI’s going to take your job,” or even “I can do your job in 3 seconds with ChatGPT.” However, most of these statements are from people who’ve been fear-mongered, and they aren’t looking at the bigger picture. If we looked at hiring rates, the types of jobs being posted, and the reason people are having trouble job searching tech market, we could shift our focus and support each other’s careers.
So, why is this such a big deal in the first place?
When ChatGPT was released in 2022, people quickly began seeing how well it could code. As a result, there were mixed reactions. Many people saw it the same way mathematicians saw the rise of digitalized math computers, and many also saw it as a total replacement for software engineers. Technically, both perspectives are valid. To reiterate the mathematicians’ view of calculator analogy, we can see that mathematicians saw calculators as a tool, and they adapted to integrate them into their daily work. Instead of letting the calculators and machinery calculations intimidate them, they instead learned to use them as a way to speed up calculations, and because of that, we’ve made so many discoveries in math, science, and everything technology-related. If many had been discouraged from their passions about math, who knows how many life-changing discoveries we would’ve missed out on.
One of the primary reasons that everyone is talking about AI taking over software engineering (SWE) jobs is the release of DevinAI. DevinAI was released in March 2024, described as an “AI software developer” capable of performing the tasks of a mid-level engineer. However, they were speaking ahead of their time, and results showed DevinAI was slow, inconsistent, and prone to errors. Despite its flashy demos, it claimed it “completed real jobs at Upwork”, or even the fear-mongering done by this, investigators took a closer look and realized it was all a sham. Although it was able to complete some tasks, the pull requests DevinAI created for new features on apps were small and usually had around 15 line changes. Something that anyone with mediocre knowledge in computer science could do. Although this is great and the development happening at DevinAI is impressive, the claims made by them and the topics people are discussing could be toned down.
Before you start panicking about how AI can supposedly easily do your job or how you’re never going to find a job, look at this graph.
Graph by FRED displaying Indeed SWE job postings in the United States.
Now, look at this graph.
Graph by FRED displaying total Indeed job postings in the United States.
You may notice a sharp decrease in SWE job postings, but if you look at the second graph, you’ll see a decrease in job postings altogether. The data for this is even stronger with other job postings like accounting, construction, etc. SWE jobs aren’t going down; merely the activity on Indeed and other hiring platforms is.
So why do I still argue to pursue computer science?
Well, the answer is that the demand for problem-solvers, engineers, and creators is not diminishing; it is merely evolving. Hiring trends will adapt, tools will update, and AI will be incorporated in some shape or form into almost every workflow. But the essence of computer science, with its logic, design, and imagination, is still required. The real challenge comes from learning how to use AI as a tool to improve your skills, rather than replacing them. You can use AI to help you debug more quickly, generate ideas and design, or handle repetitive code, but don’t let it think for you. AI can produce functions, but it won’t consider what the product is trying to produce, or deliver a designed software solution, or make ethical decisions, skills that differentiate great engineers.
To adapt, you should concentrate on learning what AI does, not just learning how to use it. Learn about model biases, prompt design, and data handling so you can use these tools appropriately and judiciously. Build initiatives that utilize AI technology, rather than relying on AI to do your work for you. Engineers who will flourish in this new environment will be those who can combine creativity and technical knowledge, who will be able to use AI to develop quicker, smarter, and more efficiently, and who the tech industry will reward.
Build impressive projects, share your journey on social media platforms, and connect with others. Good luck!
Be First to Comment