Devin 1.2: Updated AI Engineer improves coding with smarter in-context reasoning, voice integration

last year, Cognition kicked off the AI ​​agent wave with a product called Devin — the first in the world AI engineer. The proposal was kept under wraps for several months, but is now publicly available and learned very quickly. Case in point: the Scott Wu-led startup just released Devin 1.2, which brings a bunch of new capabilities to take an AI engineer’s ability to manage entire development projects to a whole new level.

The biggest highlight of Devin 1.2 is the improved in-context reasoning that makes the agent better at handling and reusing code. It also includes the ability to accept voice messages through Slack, giving users a more seamless way to tell Devin what to do.

The development comes at a time when AI-based agents are being touted as the future of modern work. Experts believe that a time will soon come when humans and agents will work together, with the former seamlessly handling repetitive tasks (which is already starting to happen). Recently at CES, Nvidia boss Jensen Huang said that in the future, enterprise IT departments will become AI “HR departments”, responsible for commissioning and maintaining agents working in various functions within the company.

What does Devin 1.2 bring to the table?

Although not a major upgrade, Devin 1.2 introduces some interesting features to make the agent better at its job. The number one feature here is the improved ability to reason in context in a code repository. This essentially means that Devin can now better understand the structure and contents of a repository.

With this understanding, the agent can identify which file is appropriate for a specific task, recognize and reuse existing code and patterns, and be more accurate in suggesting edits or creating pull requests (PRs), reducing errors and manual corrections .

For developers, this capability would mean accelerated workflows and reduced cognitive load from searching for files, understanding codebases, or fixing inconsistent code.

The other notable update with Devin 1.2 is the introduction of voice messaging. Devin can also accept voice commands from users via Slack.

All you have to do is tag Devin in a Slack chat, hit the Record Audio Clip button, and describe the task or feedback the AI ​​engineer needs to complete. Devin will prepare a step-by-step action and start executing the command using his developer tools – his own shell, code editor and browser.

This move simplifies the way a human interacts with the agent, saving the tedious typing of natural language prompts into Devin’s chatbot-style interface.

Improved login process, new corporate controls

Cognition has also made some usability improvements to Devin.

For example, in the new version, the company introduces machine snapshots to simplify the process of logging into Devin’s workspace.

“If you sign in for Devin during Devin’s browser integration, we’ll save the cookie for future sessions (if the cookie expires, you’ll need to provide Devin credentials in Secrets as well). It also unblocks authentication processes that require visiting Devin’s machine URL,” the company wrote in blog post.

Cognition is also introducing enterprise accounts, where organization administrators will get a centralized console to manage multiple Devin workspaces, including members and their access controls, as well as billing for them.

Finally, the company is adding a consumption-based billing model that allows users to pay for additional capacity beyond their subscription limits. Thus, once users have exhausted their monthly ACU allocation, they can continue to upgrade beyond that limit by paying for additional usage.

The model has been active since January 9, with users able to set their additional budgets to use according to their needs. This allows users to keep costs under control while providing uninterrupted service when they need additional capacity.

Currently, Devin is generally available for engineering support starting at $500 per month — with no seat limits. Multiple businesses are already incorporating it into their workflows, including Lumos, OpenSea, Curai Health, Nu Bank and Ramp.

Devin’s new capabilities come as competition heats up in the AI ​​engineering space. from Copilot on GitHub widespread adoption of Magic and Poolside AI raising significant funding to develop cutting-edge capabilities, the race to create the best AI Coding Assistant intensifies. Each player aims to redefine software development, promising faster workflows, reduced cognitive load, and seamless human-AI collaboration.

As these AI-powered agents continue to evolve, they are not only transforming the way developers work, but shaping the future of modern work itself, where efficiency and innovation are driven by a partnership between humans and machines.

Until 2028 Gartner an estimated 33% of enterprise software applications will include agent-based AI, enabling autonomous decision-making in 15% of daily work.