Musk’s DOGE engineers work on software to expedite federal workforce firings

Engineers working for Elon Musk's so-called Department of Government Efficiency (DOGE) are developing software to streamline mass firings of federal workers.

The software, named AutoRIF (Automated Reduction in Force), was originally created by the Department of Defense over 20 years ago. It has since been updated multiple times and used by various agencies to expedite workforce reductions.

Internal database screenshots reviewed by WIRED show DOGE operatives accessing AutoRIF and modifying its code, per Caliber.Az.

One repository in the Office of Personnel Management’s (OPM) GitHub system, titled "autorif," is located in a space designated for the director's office, which has been overseen by Musk’s associates since the Trump administration. Changes to the software were made as recently as this past weekend.

Currently, federal agencies conduct firings manually, with human resources officials reviewing employee registries and lists provided by managers. Newly hired or probationary employees — who lack certain civil service protections — are often the first to be targeted for termination. Thousands of workers across various agencies have already been let go in recent weeks. With the new software and potential AI integration, there are concerns among government employees that mass terminations could happen more rapidly.

While DOGE could use AutoRIF as it was originally designed by the Department of Defense, multiple OPM sources suspect that Musk's engineers may be developing their own software based on, or incorporating code from, AutoRIF. Screenshots show Riccardo Biasini, a former Tesla engineer and director at The Boring Company, seemingly tasked with modifying AutoRIF’s code on GitHub. One file description, authored by Biasini’s username, reads: “Remove obsolete versions of autorif.”

Biasini is also listed as the main point of contact for an email system created by the Trump administration within OPM to solicit resignation emails from federal employees.

To conduct reductions in force (RIFs), HR officials must rank employees for potential termination. AutoRIF automates this process, according to a former government HR official. "Even with any automated system, OPM guidance states that all data must be manually confirmed, and employees (or their representatives) can review the registers," the source explains. It remains unclear whether AutoRIF's functionalities have been altered by DOGE or the Department of Defense.

The news of DOGE’s involvement with AutoRIF coincides with preparations for a second major round of firings. On February 22 evening, government workers received an email from OPM requesting them to detail their achievements over the past week. Some agencies, including the FBI, instructed employees not to respond. During a meeting on February 24, OPM told agencies they could disregard the email.

The email asked workers to list their top five work accomplishments from the past week. According to a report by NBC News, this information would be used to feed into an unspecified large language model, which would assess the necessity of each employee.

Before the first round of firings, the Centers for Disease Control (CDC) tasked managers with categorizing employees as "mission critical" and submitting a list to higher authorities ahead of terminations, a CDC source told WIRED. "We went through a very, very deliberate process to identify mission-critical probationary employees, to ensure we kept those who had a real impact on the mission if terminated," the source said. "None of that was considered. We were simply given a list and told, ‘Terminate these employees effective immediately.’”

By Tamilla Hasanova