AI TOOLS MASTER

Engineers for Elon Musk’s so-called Department of Government Efficiency, or DOGE, are working on new software that could assist mass firings of federal workers across government, sources tell WIRED.

The software, called AutoRIF, which stands for Automated Reduction in Force, was first developed by the Department of Defense more than two decades ago. Since then, it’s been updated several times and used by a variety of agencies to expedite reductions in workforce. Screenshots of internal databases reviewed by WIRED show that DOGE operatives have accessed AutoRIF and appear to be editing its code. There is a repository in the Office of Personnel Management’s (OPM) enterprise GitHub system titled “autorif” in a space created specifically for the director’s office—where Musk associates have taken charge—soon after Trump took office. Changes were made as recently as this weekend.

So far, federal agency firings have been conducted manually, with HR officials combing through employee registries and lists provided by managers, sources tell WIRED. Probationary employees—those who were recently hired, promoted, or otherwise changed roles—have been targeted first, as they lack certain civil service protections that would make them harder to fire. Thousands of workers have been terminated over the last few weeks across multiple agencies. With new software and the use of AI, some government employees fear that large-scale terminations could roll out even more quickly.

While DOGE could use AutoRIF as the DOD built it, multiple OPM sources speculated that the Musk-affiliated engineers could be building their own software on top of, or using code from, AutoRIF. In screenshots viewed by WIRED, Riccardo Biasini, a former engineer at Tesla and a director at The Boring Company, has seemingly been tasked with pruning AutoRIF on GitHub, with his name attached to the repository. “Remove obsolete versions of autorif,” one file description authored by a user with Biasini’s username on GitHub says.

Biasini has also been listed as the main point of contact for the government-wide email system created by the Trump administration from within OPM to solicit resignation emails from federal workers.

OPM did not immediately respond to requests for comment from WIRED.

Got a Tip? Are you a current or former government employee who wants to talk about what's happening? We'd like to hear from you. Using a nonwork phone or computer, contact the reporter securely on Signal at makenakelly.32. In order to conduct RIFs, government HR officials are required to create lists ranking employees who may be subject to firings. AutoRIF does that automatically, a former government HR official tells WIRED. “However, even with the use of any automated system, the OPM guidance says all data has to be confirmed manually and that employees (or their representative) are allowed to examine the registers.” It’s not immediately clear if AutoRIF’s capabilities have been altered either by the Defense Department or DOGE.

The revelation that DOGE is working on AutoRIF comes as it seemingly prepares for its second major round of firings. On Saturday evening, government workers received yet another email purportedly from OPM demanding that they reply detailing what they accomplished in the last week. Some agencies, like the FBI, asked that employees not respond to the message. In a meeting with HR officials on Monday, OPM told agencies they could ignore the email.

In these emails, government workers were asked to lay out five bullet points explaining their top work achievements of the last week. On Monday, NBC News reported that this information would be fed into an unspecified large language model that would assess whether an employee was necessary.

Before the first round of probationary firings, Centers for Disease Control managers were tasked with marking workers they deemed as “mission critical” and then sending a list of them up the chain of command ahead of firings, a CDC source tells WIRED.

“CDC went through a very, very deliberate effort to characterize our probationary employees as mission critical or not, and that way we could keep those that would have real impacts to the mission should they get terminated,” they say. “None of that was taken into account. They just sent us a list and said, ‘Terminate these employees effective immediately.’”

About Us The Office of Analytics and Performance Integration (API) ensures an integrated and systematic framework for developing, understanding and measuring of a high reliability organization by providing the clinical field and leadership with analytics, tools, consultation and training to assess how VHA is performing as an organization.

What We Do API brings together, under one program office, seven functional areas that support data analytics and the quality of care within the VHA healthcare system. By combining expertise in data management, measurement, and improvement, API enables the entire VHA health system to use data to drive high-value and Veteran-centric care.

Key Functions and Activities Integrating reporting and response to identify and remedy variations in quality:

Providing customer-centric data, tools and reports that strengthen facility and system-level performance, quality of care, efficiency, resilience and enhance Veteran experience. Building enterprise analytics and improvement capabilities to foster continuous quality improvement, transparency and organizational learning. Enhancing facility performance through tailored consultation, analytics and improvement support. Leveraging shared expertise through partnerships internal (OEHRM, QPS, VISN, VAMC, ICC, HSR&D) & external (DHA, HHS, CMS, NCQA).

Chief Information Officer Michael L. Russo

Deputy Chief Information Officer for Implementation Wayne F. Lemon

Deputy Chief Information Officer for Strategy Brian C. Peltier

Chief Technology Officer Sudhanshu Sinha

Analytics, Review, and Oversight

Deputy Commissioner Gina P. Clemons

Assistant DC Sean P. Balser

Chief Data Officer Chuck Borges Office of Analytics & Improvements

Assoc. Commissioner Douglas A. Turnbull