What I like best about GitHub Copilot is how it speeds up everyday coding without feeling intrusive. It suggests code in real time as I type, and many times it correctly understands the context—whether I’m writing a function, API logic, or even test cases. This reduces a lot of repetitive work and lets me focus more on the actual problem rather than boilerplate code.

Another thing I appreciate is how well it integrates with editors like Visual Studio Code. The suggestions feel natural, and I don’t have to break my flow to search for syntax or examples. It’s especially helpful when working with new libraries or unfamiliar patterns—it gives a solid starting point which I can refine.

It also improves productivity during tight deadlines. Instead of writing everything from scratch, I can quickly generate drafts and then optimise them. Overall, it acts like a smart coding partner that saves time and reduces context switching. Review collected by and hosted on G2.com.

What I dislike about GitHub Copilot is that its suggestions are not always reliable. Sometimes it generates code that looks correct at first glance but has logical issues or doesn’t fully match the requirement. You still need to carefully review and test everything, so it’s not something you can blindly trust.

Another drawback is context limitation. In large or complex codebases, it doesn’t always understand the full picture and may give suggestions that don’t align with project-specific patterns or architecture. This can lead to inconsistent code if not handled carefully.

It can also become a bit repetitive. At times, it keeps suggesting similar patterns even when you’re trying to implement something different, which can be slightly frustrating.

Lastly, there are concerns around code quality and dependency. Over-relying on it might reduce deep understanding, especially for junior developers. It’s a helpful assistant, but not a replacement for solid coding skills and design thinking. Review collected by and hosted on G2.com.

It automatically corrected code ,installs requirements .txt file ,run python commands parallely as well as sequentially .also run commands in a virtual env if possible.it is very beneficial for coders have tons of files made and they donnot want to dig in for errors finding.just ask to GitHub copilot chat and it will make that sorted for you.Also it helps in running the total project from scratch from installing dependencies to testing project on corner test cases. Review collected by and hosted on G2.com.

Many a times when it don't find solution it not stops ,instead it go on a infinite loop checking the same condition again and again and hence wastage of cpu cycles and resources happen.also manyatimes it do not take user responses as they are said for.it automatically starts its computation if the user is simply asking to show file . Review collected by and hosted on G2.com.

Contextual Autocomplete: It suggests entire blocks of code, functions, and tests by analyzing your current file and open tabs.

Boilerplate Reduction: It handles repetitive tasks like writing unit tests, regex, or standard API calls, allowing you to focus on logic.

Natural Language to Code: You can write a comment describing what you want (e.g., // function to validate email using regex), and it will generate the implementation.

Multi-language Support: It works across dozens of languages including Python, JavaScript, TypeScript, Ruby, Go, and Java.

IDE Integration: It lives directly inside popular editors like VS Code, JetBrains, and Neovim, so there is no need to switch windows. Review collected by and hosted on G2.com.

1. Inaccuracy and "Hallucinations"

Code Quality: It often suggests code that is inefficient, outdated, or uses libraries that don't actually exist (hallucinations).

Bugs: It can generate syntactically correct code that contains subtle logical errors, requiring you to spend more time debugging than if you had written it yourself.

2. Context Limitations

Large Projects: It sometimes "forgets" logic established earlier in a file or fails to understand the broader architecture of a complex project.

Proprietary Logic: It struggles with custom frameworks or internal business logic that wasn't part of its public training data.

3. Privacy and Security

Data Training: Many users are concerned about their code being sent to GitHub's servers to train future models. As of early 2026, some users have expressed frustration over "automatic opt-in" policies for data collection.

Vulnerabilities: There is a risk that the AI might suggest patterns that include known security vulnerabilities (like SQL injection) if they were prevalent in its training set. Review collected by and hosted on G2.com.

Copilot fits seamlessly into VS Code with fast, reliable suggestions that keep my flow uninterrupted, even on larger tasks. It saves time on repetitive work, making it worth the cost. Setup is quick, and getting started feels effortless with minimal learning curve.

I also like that it gives access to multiple AI models. Review collected by and hosted on G2.com.

One downside of GitHub Copilot is that it sometimes feels a bit slower as compared to Cursor, especially when working on larger or multi-file changes.

But Copilot is much cheaper (around $10/month vs $20 for Cursor), and for day-to-day coding, it still covers most needs really well. Review collected by and hosted on G2.com.

What I like best about GitHub Copilot is how it significantly speeds up development by providing real-time code suggestions and auto-completions. It helps reduce repetitive coding tasks, generates boilerplate code quickly, and even suggests entire functions based on context. This allows me to focus more on problem-solving and logic rather than syntax. I also appreciate how it adapts to my coding style over time and supports multiple languages, making it a valuable productivity booster in day-to-day development. Review collected by and hosted on G2.com.

One downside of GitHub Copilot is that its suggestions are not always accurate or optimal, especially for complex logic or edge cases, so you still need to carefully review and validate the code. It can sometimes generate outdated or inefficient patterns, which may lead to technical debt if used blindly. Another limitation is that it may not fully understand the broader context of a large application, resulting in irrelevant or incomplete suggestions. Additionally, over-reliance on it can reduce hands-on problem-solving and learning for developers, particularly for those still building strong fundamentals. Review collected by and hosted on G2.com.

GitHub copilot is really helpful for speeding up coding and routine tasks. As someone who uses it frequently, I like how it suggests code while I'm typing and helps with small functions, syntax or repetitive parts of the code. The UI feels clean and blends well into tools like VS Code and the integrations with different IDEs make it very convenient to use. It saves time, reduces manual effort and helps maintain a smooth workflow when working on scripts or development tasks. Review collected by and hosted on G2.com.

One slight downside of GitHub copilot is that the suggestions are not always accurate so I still need to review and adjust the code instead of relying on it completely. Review collected by and hosted on G2.com.

Copilot has managed to develop into a fully agentic tool, which is great for agentic coding and development. It’s no longer just an AI assistant, and that completely changes how I can use it day to day. Support for MCP servers, skills, agnets.md, and similar pieces also makes it easier to use alongside other agentic tools.

The UI is fairly intuitive, and I like how it’s directly wired into VS Code. It doesn’t feel like “just an extension” anymore; it feels like a core feature of VS Code now.

The usage limits are also really generous considering the pricing, especially when you compare them to Claude Code, for example. Copilot clearly wins here for me by a lot. Review collected by and hosted on G2.com.

I dislike the data retention policy for Copilot coding agents and Copilot CLI. The retention period is far too long, especially given how much sensitive information is being shared, such as source code. I think they should reconsider this and make changes. It’s not that I don’t trust GitHub, but given the industry I work in, the idea that our data could be stored somewhere for extended periods of time is unacceptable. Review collected by and hosted on G2.com.

GitHub Copilot significantly speeds up development by providing intelligent code suggestions directly in the editor. It’s very effective for reducing repetitive coding tasks, generating boilerplate code, and even suggesting entire functions based on context. The integration with editors like Visual Studio Code is seamless, making it easy to adopt without changing existing workflows.

I also like how it adapts to different programming languages and frameworks. It’s especially useful when working with unfamiliar code or trying out new technologies, as it can suggest patterns and examples quickly. Overall, it improves productivity and helps maintain development flow without constant context switching. Review collected by and hosted on G2.com.

While Copilot is very helpful, the suggestions are not always accurate and still require review. For more complex or critical logic, it’s important to validate the generated code carefully. Sometimes it can suggest outdated or suboptimal patterns. Additionally, over-reliance on suggestions can reduce deeper understanding if not used thoughtfully. Review collected by and hosted on G2.com.

It is helpful for speeding up the development by providing real time code suggestions, auto completing functions and even can build complete production level website with its AI intelligence.

It improves productivity with its debugging capabilities also. Review collected by and hosted on G2.com.

Its not guaranteed that the generated code is always correct and optimized so developers still need to review code and test it.

Some ai suggestions can sometimes be irrelevant and also the ai models need pro plan pricing for most effective use. Review collected by and hosted on G2.com.