GitHub Copilot Reviews & Product Details

What I like best is how much it speeds up the boring, repetitive parts of coding.

Writing things like boilerplate code, standard API fetch blocks, or unit tests used to take up a lot of my time. Now, if I write a descriptive function name or a clear comment, Copilot usually drops in exactly what I need. Review collected by and hosted on G2.com.

The biggest issue for me is when it confidently writes code that looks perfect but is actually completely broken—what people call "hallucinations." If you aren't paying close attention, it's easy to accept a suggestion that uses a function or a library method that doesn't even exist. Review collected by and hosted on G2.com.

It helps me to prototype my idea quickly into a useable ready app within minutes saving my time, the UI is very useful and clean to use.it can be integrated within any code editor. The Ai is very useful as we can just prompt it to make our ideas come to reality and the pricing is reasonable and it performs better as it contains multiple ai models and the support is very good. Review collected by and hosted on G2.com.

It can be improved in the terms of thinking the models can be improved, like integrating more advanced useful and trending models rather than just generic models. Review collected by and hosted on G2.com.

It reduces the small and time-consuming parts of day-to-day coding. As a senior software Engineer, I find it very useful for generating boilerplate code, writing unit test skeletons, explaining unfamiliar code and with quick refactoring. It integrates smoothly with VSCode.

I, use the UI/UX for genrative AI discussions and Agent mode for reproducing bugs and finding solutions. My work performance has increased significantly. It definitely justifies the pricing and the onboarding is also seamless. Review collected by and hosted on G2.com.

I see the solutions are not always accurate or production ready. Manier times I have to carefully review the logic, starting impact, edge cases and whether it follows our project standards.

Another concern is that recently (Specifically from 2nd June, 2026) more tokens have started been consuming. So, Copilot is useful, but if pricing or token consumption becomes high, then the value needs to be justified properly. So overall, I like Copilot, but it works best as an assistant, not as a replacement for engineering judgment. Review collected by and hosted on G2.com.

The best thing about Copilot is how it keeps me in a flow state. I used to hate writing boilerplate or having to look up exact syntax for things like regex, date formatting, or complex array methods. Now I can just type a comment or start a function, and it usually gets me about 80% of the way there.

It’s not perfect—sometimes it suggests completely wrong imports or hallucinates a library—but the “tab-to-accept” muscle memory is real. Overall, it feels less like an AI writing code for me and more like a really fast pair programmer who takes care of the boring parts so I can stay focused on the actual business logic. Definitely worth the subscription if you code daily. Review collected by and hosted on G2.com.

My biggest frustration is when it’s confidently wrong. Because the autocomplete is so fast, it’s easy to accidentally hit “Tab” and accept a suggestion that uses a deprecated API, invents a function that doesn’t exist, or imports a package that isn’t even installed. That ends up forcing me to be a stricter code reviewer of my own work than I’d otherwise need to be. Also, when I’m working in a massive file, it sometimes completely loses the plot and starts suggesting variables or logic from three screens up, which pulls me out of my typing flow and breaks my momentum. Review collected by and hosted on G2.com.

What I like most is that I can talk to Copilot directly from GitHub and have it immediately start working on issues. I assign it an issue, describe what I need in plain English, and it spins up a branch, makes the changes, and opens a PR for me to review—all without me ever leaving GitHub.

It’s genuinely changed how I handle my backlog. The small things I used to put off—dependency bumps, quick refactors, adding tests, or fixing minor bugs—would just sit there for weeks. Now I assign those to Copilot and stay focused on the harder problems. By the time I circle back, there’s usually a PR waiting for me to review and merge.

One thing I didn’t expect is that it’s pushed me to write better issues. Copilot clearly works best when the task is well scoped, so I’ve gotten into the habit of writing issues more carefully, which also helps me think the work through. And since everything still happens through normal issues and PRs, nothing gets merged without me signing off. It fits neatly into the way I already work. Review collected by and hosted on G2.com.

What I dislike most is how quickly the quality drops once a task becomes even slightly complex. For well-scoped, self-contained issues it’s great, but as soon as something spans multiple parts of the codebase or requires real architectural judgment, the PRs it opens often miss the mark. I end up spending more time correcting them than I would have spent just doing the work myself.

The agent can also be slow. Waiting for it to spin up, work through an issue, and open a PR sometimes takes long enough that I’ve already context-switched to something else. Instead of helping, that delay breaks my flow.

It also burns through requests quickly on the more capable models, so I find myself rationing premium model usage and thinking about cost in a way that becomes distracting. And every now and then it confidently does the wrong thing—like ignoring an existing pattern in my repo or reformatting files it shouldn’t have touched—so I never fully trust a PR without reviewing every line. Review collected by and hosted on G2.com.

What I like most about GitHub Copilot is that it helps me write code faster and reduces the time spent on repetitive tasks. While working with Angular, .NET Core, and SQL, it provides useful code suggestions, helps with generating methods, and often gives a good starting point for solving problems. It also helps me understand unfamiliar code and improves my productivity during development. Overall, GitHub Copilot makes coding more efficient and allows me to focus more on business logic rather than spending time on routine coding tasks Review collected by and hosted on G2.com.

I have not faced any major issues with GitHub Copilot so far. Sometimes the suggestions may need small modifications to fit the project requirements, but that is normal. Overall, it has been a helpful tool in my daily development work and has improved my coding efficiency Review collected by and hosted on G2.com.

What I like most about GitHub Copilot is that it makes coding faster and less stressful. It suggests code while I'm typing, which saves me from writing the same things over and over. I use it almost every day for creating functions, fixing small issues, and writing test cases. Review collected by and hosted on G2.com.

One thing I don't like is that GitHub Copilot doesn't always understand what I'm trying to do. Sometimes it suggests code that looks correct but doesn't actually solve the problem, so I still need to review everything carefully. It can also repeat the same suggestions even after I've changed the code. Review collected by and hosted on G2.com.

I use GitHub Copilot every day while developing React and TypeScript applications at work. It helps me write reusable components, generate unit tests, explain existing code, and compare different implementation approaches before making decisions. I also use Copilot Chat for planning tasks, writing documentation, and speeding up repetitive development work. Review collected by and hosted on G2.com.

Sometimes the generated code is not accurate for complex project-specific requirements, so I need to review and modify it. It can also lose context in long conversations, and responses may become slower when working on large codebases or multiple files. Review collected by and hosted on G2.com.

Honestly, the autocomplete is what got me hooked. I was skeptical at first like, how good can it really be? But within the first week I noticed I was spending way less time writing boilerplate and more time actually thinking through problems. It picks up on what I'm trying to do really fast, sometimes finishing a function before I've even thought through the logic myself. The VS Code integration feels completely natural, no friction at all. And it works across languages, which matters a lot for me since I jump between Python and TypeScript constantly. Review collected by and hosted on G2.com.

Not a whole lot honestly. Sometimes it suggests something that's totally off-base like it'll autocomplete in the wrong direction if I have a particularly unusual pattern in my codebase. You still need to pay attention and not just accept everything blindly. Also the subscription cost feels a bit steep if you're just using it solo, though for teams it's easy to justify. Oh, and occasionally it lags a bit when I'm working with very large files, but that's minor. Review collected by and hosted on G2.com.

As a QA Engineer working on automation frameworks at my company, the best part about GitHub Copilot is how it accelerates test automation by generating boilerplate code, edge‑case scenarios, and automation scripts far faster than writing everything from scratch. For example, while building Pytest suites for accessibility checks, Copilot suggested assertions, mocks, and fixtures that aligned perfectly with our framework, reducing setup friction and saving hours of manual effort.

Another major advantage is Copilot’s ability to use project context intelligently. When I’m working within our company’s test repositories, it surfaces alternative approaches — such as different ways to structure parameterized tests — so I can compare options and choose the most robust one for production pipelines. This context‑aware assistance ensures that the generated code is not only syntactically correct but also relevant to the specific QA workflows we maintain.

Overall, Copilot doesn’t replace my QA judgment; it strengthens it by cutting down repetitive effort and freeing me to focus on quality analysis, exploratory testing, and ensuring reliability across systems in a real workplace environment. Its consistency and context awareness make it a reliable partner in professional QA projects. Review collected by and hosted on G2.com.

From a QA perspective in my company’s automation projects, the biggest drawback of GitHub Copilot is that it can generate code that looks polished but isn’t always dependable, especially around edge cases or complex test scenarios. For example, while working on compliance‑driven Pytest suites logged in with my company account, Copilot often suggested repetitive or shallow test cases instead of highlighting coverage gaps. This meant I still had to spend time validating, adjusting, and refining its output before it was production‑ready.

Another challenge is that Copilot doesn’t consistently grasp the deeper intent behind QA work — such as robustness, negative testing, and regulatory compliance. Its suggestions can overlook the quality checks that matter most in enterprise pipelines. In day‑to‑day use, Copilot does speed up routine coding tasks, but it requires ongoing oversight to prevent subtle bugs and to maintain consistent standards across the entire test suite.

A concrete case: during a recent sprint, I used Copilot to generate parameterized test functions for API validation. While the code compiled correctly, it missed certain negative scenarios required for compliance. I had to manually supplement these cases to ensure full coverage, which slowed down the evaluation process.

Overall, Copilot’s cautiousness and occasional lack of depth mean that while it accelerates routine tasks, it still demands careful QA oversight to ensure reliability in professional environments Review collected by and hosted on G2.com.