We use github copilot as a coding assistant mainly. We have been using the IDE integrations of Github Copilot in both IntelliJ IDEA as well as VSCode. It has acted as a very strong coding assistant, providing valuable suggestions and feedback to the code. It also suggests code for problems that I describe, thus helping me "vibe-code" in my work. We have not yet procured the agentic capabilities for github copilot so my review is based on the non-agentic version.
Pros
Reading codebases and understanding the same.
Providing valuable code snippets for problems or tasks that I describe in English.
Also helps with other supplementary technical activities, such as, from time to time, I have asked it to suggest places where I can add logs and suggest how I can monitor those logs on Splunk via a dashboard. It has helped me with both.
Cons
Sometimes code generated by Copilot is not of the best quality, doesn't handle all the edge cases, and misses some requirements.
I have noticed that the Copilot is not the best at analyzing large monolithic codebases and placing them in their context. It has suggested to me APIs and methods from time to time that don't exist in my codebases.
I would have loved it if there were a deeper integration available with the IDEs. Although the IntelliJ IDEA extension is decent, I would have loved it if there were, say, a direct integration available of the GitHub Copilot agent with the IntelliJ Debugger.
Likelihood to Recommend
Copilit is fantastic at the following: 1. Solving simple, well-defined problems, such as implementing an algorithm, manipulating a data structure, or string manipulation and regex. 2. Implementing simple APIs that are mainly CRUD in nature, with moderate business logic inside them, which may involve some processing or passing the data through an algorithm. 3. Implementation of well-defined activities, such as implementing a connection to an Oracle DB using Hibernate or JDBC, or implementing boilerplate code for a backend service to listen to Kafka events. It is not that great when it comes to understanding and implementing code in a proprietary DSL. It struggles when implementing a major feature across a complex codebase. I believe developers should also adopt the trust-but-verify paradigm when expecting highly secure or regulated code from GitHub Copilot.
We use GitHub Copilot in our organization primarily for software development especially for code generation and code suggestion, its saves lots of time and helps us to develop our app faster with efficient code. Nowadays it become more powerful with the agent mode, now it has a capacity to perform very complex task with high success rate, if efficient prompt is provided.
Pros
Real time code suggestions are really fast and accurate I liked it.
After introduction of Agent Mode now it can handle very complex task with high accuracy.
Get better performance, if use it with MCP server.
Cons
The only issue I faced is the context token size, it can be increase.
Sometime with agent mode it takes too much time, which can be reduced but I understand it also depends on the complexity of the given task.
Likelihood to Recommend
Based on our experience it is best suited for us to develop Apps faster with the help GitHub Copilot, with its smart code suggest we can build our Apps 3x faster then before. It is also suited very well for the quality code generation. We can also test security of our Apps with agent mode which helps us lot to make our app more secure. For small code generation it is perfect, but if we want to build a large amount of code at once, then it may not that accurate.
Development teams in our organization use GitHub Copilot to improve productivity while working on .NET applications in the Visual Studio IDE. Before using GitHub Copilot, our developers depended on Google searches, AI tools, and developer forums to get help with development tasks. With GitHub Copilot now integrated into Visual Studio, they have a one-stop solution that eases their development process.
Pros
Code Completions.
Fixing Bugs in Code.
Optimizing Code.
Cons
Chat responses can be improved.
Accuracy needs improvement.
Ease of Use can be made better.
Likelihood to Recommend
GitHub Copilot is beneficial when writing .NET code and needing on-the-spot assistance with development tasks such as code suggestions, bug fixing, and code optimization. It is also helpful when you are new to a technology and want to learn and use it faster. However, it might not be appropriate where you are dependent on GitHub Copilot. You can use it for assistance, but not as a replacement for your understanding and logic.
VU
Verified User
Team Lead in Engineering (Computer Software company, 11-50 employees)
GitHub Copilot is employed in our organization to accelerate development
by offering AI-powered code suggestions. It reduces coding errors,
supports various languages, and aids in learning and training. It's
particularly beneficial for pair programming and collaborative efforts,
and facilitates rapid prototyping. Copilot seamlessly integrates into
existing workflows, potentially automating code generation and assisting
with documentation. Its scope of use varies based on project needs,
from extensive adoption across all projects to targeted application for
specific tasks or projects.
Pros
Autocompletion
Suggesting code based off of comments
Fleshing out ideas based off of notes
Cons
Speed, sometimes it's a little slow
Better documentation
Ability to refine feedback after initial prompt
Likelihood to Recommend
GitHub Copilot excels in scenarios demanding rapid prototyping or
proof-of-concepts, where speed is paramount. It shines when tackling
boilerplate-heavy tasks, like setting up authentication or handling file
I/O. It's less suited for highly sensitive or critical code, where human
scrutiny and manual verification are essential. Additionally, in novel,
cutting-edge technologies where well-documented patterns are scarce,
Copilot's suggestions may be less reliable.
VU
Verified User
Engineer in Engineering (Computer Software company, 501-1000 employees)
