GitHub Copilot, an artificial intelligence assistant for writing code

GitHub presented few days ago a new function called «GitHub Copilot»Which should make life easier for programmers and as the name of this function suggests, it is in charge of reviewing the code with you, that is, it offers a smart wizard capable of generating standard constructs when writing code.

The system was developed in cooperation with the OpenAI project and uses the OpenAI Codex machine learning platform, trained in a wide variety of source codes hosted in public GitHub repositories.

Today, we are releasing a technical preview of GitHub Copilot , a new AI pair programmer that helps you write better code. GitHub Copilot extracts context from the code you're working on, suggesting full lines or full functions. 

GitHub Copilot differs from code completion systems traditional for the ability to form quite complex code blocks, up to ready-to-use functions synthesized taking into account the current context. As Copilot is an AI function that has learned through several million lines of code and it recognizes what you are planning based on the definition of a function, etc.

For example, if you want to create a function that tweets, Copilot will recognize it and suggest the code for the entire function, because there have certainly been enough programmers before that have already written such a function. This is useful because it saves you the trouble of looking for examples in other code snippets.

It helps you quickly discover alternative ways to solve problems, write tests, and explore new APIs without having to tediously tailor an Internet search for answers. As you write, it adapts to the way you write code, to help you complete your work faster.

Another example, is if there is an example of a JSON structure in the comment, when you start writing a function to parse this structure, GitHub Copilot will offer out-of-the-box code, and when the user writes repeating description routine enums, it will form remaining positions.

With this we can understand that GitHub Copilot it adapts to the way a developer writes code and takes into account the APIs and frameworks used in the program. 

According to GitHub, it is "significantly more capable than generating GPT-3 in code generation." Because it has been trained on a dataset that includes more public source code, OpenAI Codex should be more familiar with how developers write code and be able to submit more accurate designs.

For those who are interested in being able to try Copilot, you should know that it can be integrated into Visual Studio Code as an extension and it goes far beyond just completing a command. Preview officially supports code generation in Python, JavaScript, TypeScript, Ruby, and Go programming languages, but it can help with other languages ​​as well.

OpenAI Codex has extensive knowledge of how people use code and is significantly more capable than GPT-3 in code generation, in part because it was trained on a data set that includes a much higher concentration of public source code.

In the future, it is planned to expand the number of supported development languages ​​and systems. The work of the plugin is done by calling an external service that runs on the GitHub side, to which, among other things, the contents of the file edited with the code are transferred.

Finally, it is worth mentioning that the concept of something that is actually automatic code completion based on artificial intelligence is not entirely new, since for example Codota and Tabnine have been offering something similar for a long time, in addition to combining their activities and last month they agreed to Tabnine as the main brand.

We can also mention Microsoft which recently introduced a new feature, Power Apps, which uses the OpenTI GPT-3 language model to help users choose the correct formulas.

If you are interested in knowing more about it, you can check the details in the following link.


Be the first to comment

Leave a Comment

Your email address will not be published. Required fields are marked with *

*

*

  1. Responsible for the data: Miguel Ángel Gatón
  2. Purpose of the data: Control SPAM, comment management.
  3. Legitimation: Your consent
  4. Communication of the data: The data will not be communicated to third parties except by legal obligation.
  5. Data storage: Database hosted by Occentus Networks (EU)
  6. Rights: At any time you can limit, recover and delete your information.