The world's first AI software engineer, ensuring whole team work

Recently, technology startup Cognition Labs introduced the Devin tool, the world's first software engineering AI.

This artificial intelligence (AI) tool has advanced reasoning capabilities and automatically takes care of many software development tasks. This is considered a new step forward in artificial intelligence, but it also makes many software engineers worried.

As we know, language models like ChatGPT 4, are capable of writing programming code, Devin - a new AI tool from Cognition Labs - can do much more.

This is not a simple on-demand coding tool, it is a virtual software engineer, which can extend its capabilities to create other AI models.

This new tool makes many scientists really admire, but it also raises many concerns for the software programming industry. Will we still need programmers in the future?

This tool can guarantee the work of a team of programmers. (Illustration photo: Trust my science).

Cognition Labs has tested its product in real-world tasks.

In a test, engineers tasked this AI tool with setting up, debugging, and testing a computer vision model.

Ensure the role of the whole group

Devin can do the work of an entire team of software developers. It ensures completion of all stages, from planning, defining requirements to implementing the software design.

The engineering team integrated project management and business analysis skills, allowing Devin to understand the goals and plan the steps needed to complete the project.

To complete the task, Devin creates "child assistants" that take on part of the work, they interact with different development tools (terminal, code editor, browser). to code, test, and debug software.

In addition, Devin also has the ability to learn quickly and adapt to new technologies. All it takes is reading blog articles to understand how a tool works and then using it.

Create AI and solve problems

In the test video, the Devin tool copied Meta's Llama language model, read the README file to understand how to configure it, then automatically started a training cycle.

In the above test, the Devin tool removed or fixed the inactivity error. This proves that it is capable of solving programming problems independently.

In another test, Cognition Labs wanted to evaluate Devin's performance in solving deeper programming errors.

The engineering team used open source projects found on GitHub, and the above tool resolved nearly 14% of the identified bug sets without any outside help.

What does this technology mean for developers?

Devin's appearance and the rapid advancement of AI in general are certainly causing concern for software developers.

However, at the present time, companies cannot completely entrust their software and application development to AI, especially companies that require high reliability and security.

The idea of ​​allowing AI to "infiltrate" entire code and make automated edits will certainly also raise a wave of concerns about the security aspect.

For example, changes to AI could introduce bugs or security vulnerabilities, especially if the AI's decision-making process is not transparent.

Therefore, we will still need time and reliable evidence for business customers to fully trust this type of technology.