“It’s Still Evolving” and “I’m Smarter Than Copilot”

Yes, you are, but Copilot can still be incredibly useful.

Many people overlook this, but software development is more about thinking than typing code. When building something for widespread use, most of your time goes into figuring out the best way to design it—not just writing lines of code. And that’s exactly how it should be.

As engineers, our primary goal should be to maximize the value users get from our products. We should focus on creating an experience that’s both enjoyable and valuable, rather than getting bogged down with naming variables or selecting libraries for validation.

This is where GitHub Copilot shines. It can handle routine coding work, freeing you to tackle the bigger challenges.

Think of it this way: you’re a senior engineer with skills Copilot doesn’t have, while Copilot is great at generating code quickly. When you shift your perspective from “Copilot doesn’t know enough” to “Copilot can help if I guide it correctly,” it becomes a much more effective tool.

In software development, every feature is about 70% planning and 30% coding.

In my own quest to increase productivity, I’ve incorporated Copilot into my workflow and compiled a few tips that have worked well for me.

Provide Structured Context

Give Copilot a clear view of your data structure. This can significantly improve the quality of code it generates. We often underestimate how much we rely on structured data, but when Copilot has access to your schema, it can produce much better code for manipulating that data. Remember, Copilot is a generative AI—better context leads to better results.

For example, if you’re building a shopping cart and need CRUD endpoints, provide your product data structure. Copilot can then generate all endpoints for you. Afterward, select the code and ask it to add validations—it’ll handle that as well.

Delegate All Routine Coding

Once you’ve set up your data structure, ask Copilot to write the functions for data manipulation, fetch database results, and format the output as needed by your application. By offloading these repetitive tasks to Copilot, you’re free to focus on higher-level engineering work.

Use Copilot in Small, Targeted Tasks

It’s easy to fall for marketing hype about a product, and then feel disappointed when it doesn’t meet every expectation. Copilot is not here to replace you—at least not right now. Where it truly excels is in handling specific use cases with a lot of context. For these tasks, it can be incredibly helpful.

With the right perspective and guidance, Copilot can be an excellent productivity booster, handling the routine work so you can focus on the creative and strategic elements of software development.