Sahil Lavingia, well-known Silicon Valley entrepreneur and founder of Gumroad, recently recounted his brief, 55-day stint with Elon Musk’s Department of Government Efficiency (DOGE), a temporary agency created by President Trump via executive order. In a personal account detailing his experiences, Lavingia shared insights from his time as a volunteer engineer at the Department of Veterans Affairs (VA), where he joined DOGE to help streamline the massive agency’s technology infrastructure.
Initially tasked with identifying potentially wasteful contracts and candidates for layoffs, Lavingia quickly discovered that personnel rules within the VA were more nuanced and complex than anticipated, including seniority and veteran status considerations that were prioritized over employee performance alone. Further challenging his assumptions, he noted that despite widespread beliefs about inefficiency, the VA was surprisingly functional, albeit slow-moving.
Lavingia described DOGE’s role at the VA as akin to that of external management consultants—advisory rather than authoritative. He highlighted that real decision-making powers remained with the agency heads directly appointed by President Trump, suggesting DOGE primarily served as a convenient scapegoat for unpopular moves, a dynamic Elon Musk himself publicly lamented in recent interviews.
Despite his enthusiasm to leverage his Silicon Valley background to innovate within government services—such as improving the UX of VA applications and accelerating veterans’ claims processing—Lavingia expressed frustration that he never received approval to deploy significant changes that could tangibly benefit the public or reduce taxpayer costs.
Nevertheless, Lavingia produced several open-source projects during his brief tenure, including tools designed for document analysis using large language models (LLMs), contract evaluation software, and organizational chart generators. One notable creation was a program capable of scanning internal documents for references to diversity, equity, and inclusion (DEI), gender identity, COVID-19 policies, and partnerships with organizations like the World Health Organization.
Lavingia also criticized DOGE internally, remarking on its apparent lack of structure and failure to maintain centralized knowledge-sharing resources, which caused inefficiencies as each engineer entered the program “from scratch.”
His tenure at DOGE ended abruptly after he shared details of his work with the press. Lavingia found himself dismissed without prior notice, his access revoked immediately after a Fast Company interview was published.
Reflecting on his experience, Lavingia expressed appreciation for his firsthand exposure to the complexities of governmental bureaucracy. Although disappointed at the limitations of his role and DOGE’s organizational shortcomings, Lavingia ultimately acknowledged that federal operations, while slow, do function more effectively than he had anticipated. His brief journey underscored the inherent difficulties of rapidly applying Silicon Valley approaches to large-scale institutional reforms within the entrenched structure of government agencies.
