GitHub CoPilot Reviews

Whenever I need to write test cases for my development code, I just give the command to Copilot, and it automatically helps me write them. Then I just need to review them. It also really helps me write the skeleton of the basic design. It won't give me a 100% solution, but it gives me a 75% solution. Then I just need to review it, make some cosmetic changes, and it's ready. 

Overall, where I used to spend a whole day on test-related activities, now I just need to spend two to three hours, and I can complete my job.

It's helped reduce engineering effort. If I can complete my job in half a day, then all the engineers can complete their jobs in half a day. Ultimately, it saves cost and optimizes performance. 

From the company perspective, if we needed to spend a million dollars, we can now reduce it to at least a hundred thousand dollars.

Copilot's code review, test case generation, and chat features are most valuable to me. When I write code, I feel like I have someone sitting with me to help me. Wherever I'm stuck, I just ask a question, and it provides guidance that I can use.

In a few cases, the results aren't correct, so that needs improvement. Also, it would be great if the results could be presented in different formats, not just text. As engineers, it takes time to read through text-based results. 

If they could provide diagrams or basic design architectures, it would be easier to understand and take less time to review. So, as of now, image functionality is not provided by GitHub Copilot, and it would be great if they could add that.

So, I would like to see improvements in the result generation, more results, different result formats, and image functionality.

I have been using this solution for four months. I work with the latest version.

I would rate the stability a nine out of ten. It is a stable product. Copilot is a good thing when you're writing code. It assists you wherever you need it. 

On the other hand, there are a lot of benefits to this feature, like suggesting code before you write it and suggesting the optimal way to write it, which is really helpful. 

Apart from that, it helps with writing test cases and suggestions for heavy algorithms where you need to put in a lot of brainpower. So, it makes things easier.

On the other hand, the negative points, Copilot still hasn't reached a point where we can trust it 100%. Sometimes, when we raise a query, it stops working intermittently. So, at that moment, we need to look for other options. It's not, as of now, 100% reliable. Sometimes, it stops working; sometimes, the output has issues. So, we need to be careful with information.

I would rate the scalability an eight out of ten. Sometimes, we face high traffic on Copilot, and it stops responding to our queries. It's happened a few times, maybe three or four times out of a hundred.

It's not 100% reliable. Sometimes, we stop receiving responses. We have around 500 end users in our company. It is used on a daily basis. 

The customer service and support are okay. 

Positive

We switched to GitHub because it's more appropriate and saves me time. On Google, I need to spend time searching for something, typing it in, and then getting a bunch of solutions out of which I need to pick the best one. 

Copilot, on the other hand, gives me the best answer out of all the possibilities with just one query. That's a big benefit.

I would rate my experience with the initial setup of this product a seven out of ten, where one is difficult, and ten is easy to set up. 

It is not complex. I gave it a seven because there were a few minor technical challenges with the initial login. It wasn't as smooth as logging in to Gmail or Microsoft Outlook, but I was able to log in after a few tries. 

So, it is a bit tricky. These are just minor challenges, not completely smooth. 

We use the licensed version, not a cloud version. We purchased a license and installed it.

We did everything ourselves. I was able to do it alone; it didn't require much effort. So, one person can manage both the deployment and maintenance.

It took only one day to deploy it. It's not really a deployment in the traditional sense. We requested a license key from Microsoft, and then we used that key to log in from our laptops. We configure it with our development environment, and it works smoothly.

If you invest some amount, it will definitely reduce your production cost. I would rate the ROI an eight out of ten, where one is low ROI and ten is high ROI. 

I would rate the pricing a five out of ten, where one is very cheap and ten is very expensive. So it's affordable.

There are no additional costs to the standard license. So the price is fixed.

This is the first time I've used a tool like this that assists with code writing. Before that, I used a simple Google search. 

Moreover, Google has recently launched its feature, but it's not as mature as Copilot's. They're in the very early phase.

Developers, testers, front-end, back-end, any developer can use it. It's really beneficial for them. On the other hand, testers can also explore test cases with Copilot. It'll help them, too.

Overall, I would rate the solution an eight out of ten. 

The primary focus has been on leveraging its capabilities for increased ease of use and productivity. The main goal is to accelerate application development, allowing us to build applications more efficiently.

The most valuable aspects in terms of features, revolve around its code generation capabilities. Firstly, the functionality for generating test cases based on existing code and application logic has been invaluable. Secondly, the code explanation feature has emerged as another essential tool. Beyond these, the primary function of facilitating the development of new code or additions to existing code remains a key focus.

There's room for improvement to ensure that suggestions align more precisely with the context of what I'm seeking, minimizing instances of unrelated or inaccurate code suggestions. It could benefit from enhancements in several key areas. Firstly, there's a need for improved integration across different Integrated Development Environments to ensure a consistent user experience. Secondly, providing users with more granular control over when Copilot is active, preferably on a per-project basis, would enhance flexibility. Also, broadening Copilot's contextual awareness beyond open files to consider the entire project or codebase context could lead to more accurate and relevant suggestions, particularly in scenarios where opening all files is impractical. The current state lacks maturity, and reporting is not readily available out of the box. Users have to request specific reports, involving collaboration with the data team, indicating a need for more streamlined and accessible reporting features.

I experimented with the technical preview version when it was initially launched last year, taking advantage of the free access provided at that time. However, in a professional context, I have been utilizing it for real-time database work over the past three and a half months.

The stability is generally satisfactory. The tool itself operates smoothly, but potential challenges may arise, especially in larger enterprises utilizing proxies. It's crucial to adhere to the documentation on operating with proxies to avoid glitches.

The scalability is impressive. It accommodates growth seamlessly; as you acquire more licenses, the tool scales accordingly to meet the increasing demand. As of now, we have successfully onboarded and currently have approximately three hundred and fifty active users.

We have engaged with the support team for feature requests, particularly in terms of porting, and have found the response to be somewhat delayed. It's worth noting that GitHub and Microsoft teams have communicated that they are actively addressing these concerns in private.

The initial setup is straightforward.

The initial setup operates on a license-based model, providing flexibility as you can allocate licenses to a GitHub team. While the implementation itself is straightforward and requires minimal ongoing maintenance, a significant challenge has emerged in terms of documentation, especially concerning security documentation. The current lack of comprehensive documentation raises concerns, particularly for larger enterprises. When activating Copilot, you are prompted to provide essential information, such as the location of your Azure instance where suggestions will be processed. Once you input this information, the installation proceeds seamlessly.

While it's premature for a comprehensive ROI analysis, we have observed a positive trend with users expressing satisfaction.

A personal license is priced at ten dollars per month, while a professional or enterprise license costs nineteen dollars per user, and these rates are consistent for all users.

While we have explored various options within the Microsoft ecosystem, such as the solution offered, we haven't actively employed any other AI tools at the enterprise level.

It's essential to perceive it as a supplementary tool, not the primary driver. Caution should be exercised in blindly accepting suggestions to avoid unexpected challenges. The time invested in reviewing suggestions is crucial, and the tool's effectiveness depends on the user's domain knowledge. While it can be ideal for bridging domain and technical aspects for those with expertise, users with less experience may overlook potential issues, necessitating careful consideration. Overall, I would rate it seven out of ten.

GitHub CoPilot helps its users summarize documents. GitHub CoPilot has an AI capability that allows it to give its users a summary of documents that it has received from them. GitHub CoPilot also provides meeting summaries and suggestions on how and when you are supposed to send reports or emails. GitHub CoPilot establishes a total baseline of its users' activities so that it can later recommend to its users how to structure their day.

GitHub CoPilot's stability is an area of concern where improvements are required.

I have been using GitHub CoPilot for three or four months. I have experience with the preview version of the product. Though I am not sure about it, I think that GitHub CoPilot is not yet released in the market.

Stability-wise, I rate the solution a seven out of ten.

I will not be able to comment on the scalability part of the solution since my company has not undertaken a full-fledged deployment process across our organization's users.

All the people in my company use GitHub CoPilot in one form or the other.

The product's initial setup is straightforward because it's pushed through the updates. The product's updates are allowed automatically, so the setup process is not something one needs to get involved in deeply.

The solution can deployed in less than a minute since it is done through the automatic updates provided by GitHub.

No technical people are required to take care of the deployment and maintenance of the solution since it gets automatically pushed as an update from GitHub CoPilot.

The solution's deployment part was managed with the help of one of our company's in-house teams.

GitHub CoPilot comes readily available for enterprise customers, so it's a free add-on if you already have GitHub's enterprise license.

I recommend the solution to those who plan to use it.

As it is a new product in the market, I rate the overall tool an eight to nine out of ten.

Today, if you're aware of the CI/CD requirements, then it's fine. Unless you know how to create the pipeline, it would be tough. The one I'm using is already set. So I can implement it immediately. If there is a deviation from that, then it will be a little tough, I would need to Google to find the solution. I can implement it.

When you're using it, it's automatically deployed. In a normal application, it used to provide a single application. In this proprietary thing, it is just now it used to provide in the scripting tag, you need to provide Spring Boot.run, and that server code. So that unit is forwarded. In Copilot, you need to mention which nodes you are going to configure. So you need to provide all the port numbers and then deploy there. It will automatically deploy everywhere.

The deployment is easier when you're handling multiple deployments. So, I don't need to node application unit for node, update, and everything everywhere. But it's also in the engine we can do there. 

Once you set up everything, it is automatically handled. But once the setup is done, it's done.

GitHub Copilot code suggestion capabilities: It's always good. When you're doing any deployment, the suggestions are always welcome. You don't need to remember everything. You can pick from the solutions provided.

One thing I noticed is that normally, in CI/CD pipelines, we put: 

• the Jenkins path, 
• the Git repository paths, and 
• the number of reports we are going to deploy.

If we had an automatic system, an automatic page to do all these things, it would be easier. Not a configuration, but a page. If you have that kind of page, then Copilot can help all the developers. This is my individual preference. It can be there, it may not be. Maybe I'm not able to explore those issues, but that is one thing I noticed.

Currently, we need to map everything. We need to provide the Git card and where we need to deploy. If it is pre-prod, stage, or production, we need to map it separately. If we had some kind of program and could take it from the application of properties file, it would have been great. We configure everything in separate files. 

Even for development, staging, and production environments, we need to make separate files for each. If you have some mechanics to automate that, it would be great. It's for the entire system. Their system is like that. You need to prepare separately for every manage. So if you have one utility to make all those alternates, it would have been great. A UI would enhance the developer experience and streamline the CI/CD process.

You need to configure it separately for each environment. If everything is in a dropdown, it could be directly included. It will automatically connect to your dedicated agent. But it is okay. As of now, it's working fine. It's just manually maintaining that data. It worked fine. No issues. That is the only suggestion I have.

TL;DR:

For different environments, like production, stage, and development, if we can have a single point, like one, two, or three instead of opening for each, that feature would be helpful. We have numbers for certain regions, but again, the developer and the team should remember which number is for dev, stage, and prod. That could be easier for maintenance. 

The administrator doing the CI/CD pipeline should remember that number. It can be easier for any AI tool to do that. Instead of putting the phase in the configuration, if you put in a AI tool, it is easier for the system to manage.

I have been using it for three months. 

It's stable. 

It's scalable.

In the last team that we had around 13 members, everyone never faced any issues when using Git merge and Git commit. You have a separate Git repository, and you just need to know how to use it. The only thing is, our case of sessions should be organized, and the resources should be proper.

You want to see what everything the account does. I do everything with a command prompt, not using any Git GUI. You'll get the GUI application for your commit and such. It's fine.  

Deployment time: If the database scripting and the data center information are ready, then it will take one or two hours. It's a server configuration. Two to three hours of work.

Deployment resources: I can also do it. It doesn't require multiple people. If the software is ready and everything is fine, then only one person can do everything. That is what a full-stack developer means. If you are a full-stack developer, it means you can handle all the data-related tasks and services. I'm a full-stack developer, so I'm ready to do that. I should have the capability to do everything on the website.

We have a separate team to do that. They will do the things for us. From my side, I just raise a ticket in Jira support, and then they will take the ticket and implement it. The installation and configuration is a separate process because in each organization, they have separate monitoring systems.  

There's a separate team to make those things happen for me.

Overall, I would rate the solution a nine out of ten. The only suggestion is to enhance Copilot's ability to assist developers with infrastructure as code tasks.

Like, while using CI/CD pipeline, when I use YAML files, so it could just support the port number. That will improve the feature. But the developer should be aware of the feature.

We use the tool for code snippets that we integrate into our solutions. 

GitHub CoPilot helps us auto-complete code. The tool's context awareness benefits our projects because it identifies the context in which we are working and automatically proposes solutions that we use. 

The most important factor is that we need less time for design. We can cover solutions more quickly; our updates will be within days, not months, as before.

The tool needs to focus on integration, as it is the most important aspect. I would like to see some pre-designed modules included in my projects.

I have been using the tool for a month. 

I rate the solution's stability a nine out of ten. It is good and promising. 

GitHub CoPilot's scalability is quite satisfactory. I rate it a ten out of ten. My company has two users. 

We use support from Microsoft's online documentation and resources. We don't need to contact human support.

GitHub CoPilot's deployment is straightforward. It is used as a tool for assistance rather than being deployed within our projects.

We have a demo license. Once we understand what we'll do, we'll start with a paid license.

We are still in the planning phase. We have attended some seminars by Microsoft to understand our needs better. Once we have a clear plan and strategy, we will discuss whether we need professional help. 

My advice for people who want to start using it is to understand how it works and to find uses for the platform. This means focusing on how to develop their processes, not traditionally, but by utilizing these AI tools.

I rate the overall solution a ten out of ten. 

Most of the time, it is for code suggestions. It allows improving the efficiency and productivity of development. It is used by developers within the IDE, so the development environment.

Not that much, as I need to wait until the ongoing pilot ends to have final feedback. However, it is expected to improve coding productivity and the capacity to produce quickly and more efficiently.

The most known and important features from a development point of view are the code suggestions. It enhances efficiency and productivity in development. 

The merge request features and workflow are available. We don't use them much currently. GitHub Copilot enables the production of business projects quickly and efficiently.

I cannot really answer this one as I don't have enough experience to share this kind of feedback. Someone who uses it daily might better understand these questions.

I have used it for around six months.

Very high stability. I didn't notice any issues regarding stability.

The AI model is really scalable, so I would rate it at least nine out of ten. I am optimistic about the next releases being developed and designed.

I do not have experience contacting technical support, so I cannot provide feedback on this aspect.

Positive

Before GitHub Copilot, we evaluated GitLab with its Duo feature. Developers chose Copilot because of stronger code suggestions.

The initial setup was easy, particularly within the IDE. Minimal complexity was involved.

We use GitHub Copilot internally with the client's company. It is an internal project, and as an architect, I was involved in the discussions to decide between Copilot and other solutions.

I need to wait until the pilot we are undergoing ends to get a clear understanding of the ROI and productivity improvements.

I am not really included in the cost management part, so I don't have insights into the pricing or setup costs.

The other option considered was GitLab Duo. Based on presentations and evaluations, we chose GitHub Copilot due to its stronger code suggestion capabilities.

I would initially rate it nine out of ten based on what I know and the small experience I have had. For it to be ten out of ten, it needs more advancements in the AI part, though it is already strong. Future updates are anticipated.

I'd rate the solution nine out of ten.

We use the platform to accelerate coding by providing relevant code suggestions and automating repetitive tasks. I use it in a development environment focused on various programming languages, where it helps by predicting code snippets and detecting variable names, which speeds up the coding process.

The platform's most valuable feature is detecting and suggesting variables based on the existing file context. This feature streamlines coding by offering relevant variables, making it easier and faster to write accurate code.

They could enhance the product's accuracy of suggestions, especially when dealing with non-standard or legacy code. Additionally, an easier and more intuitive way to toggle Copilot on and off would be beneficial.

I have been using GitHub Copilot for a few months.

The code prediction is generally stable, though there are occasional inaccuracies. 

The initial setup was very straightforward. 

The implementation was done in-house.

The return on investment may vary depending on the project. The predictions aren't always accurate for large enterprise solutions or extensive codebases. However, for standardized projects, it can significantly improve productivity.

GitHub Copilot is a paid service, costing approximately $6 per month. There might be a free trial available.

I decided to try GitHub Copilot because of its integration with GitHub and my existing workflow.

GitHub CoPilot is quite capable of AI. It provides good suggestions and aids in coding, but the increase in productivity is modest.

I recommend that others analyze if it works for their requirements. It's important to assess if it fits their workflow before fully integrating it into their coding practices.

I rate it a seven out of ten.

GitHub CoPilot can be used for prompt engineering, delivering information to create a presentation, and comparing results from other AI tools, like ChatGPT or Gemini. You can use the CoPilot app while creating presentations, but we need to subscribe and pay an additional fee to use it. I'm truly considering talking to the management team to contact Microsoft to see if we can use that feature. It's a very useful feature.

The CoPilot app that you can use while creating presentations is very useful.

It would be useful if we could use the solution to create documents, not only prompts. It would also help if we could insert documents and create results in accordance with the documents inserted.

The tool has a feature that allows you to upload a voice document or a recording, and it can export the transcription for that recording. This transcription is not available in all languages. I would like to know whether the solution plans to expand the number of languages because I want to know if there's also a plan for Hebrew transcription.

There are limitations because you have to subscribe and pay more money to use the GitHub CoPilot tool.

I have been using GitHub CoPilot for a few months.

I rate the solution an eight out of ten for stability.

I rate the solution’s scalability an eight out of ten.

I've tried to contact Microsoft, but I haven't received a response. I'm not impressed with the technical support.

The solution's initial setup is pretty straightforward, but I'm not as impressed as ChatGPT and Gemini.

With GitHub CoPilot, we have more opportunities and can approach more people in less time. The solution shortens the time it takes to finish a certain task.

We have a license but need another one for the GitHub CoPilot tool.

Overall, I rate GitHub CoPilot an eight out of ten.