Doge Plans to Rebuild SSA Codebase in Months

Practical Concerns from Beneficiaries and Caregivers

  • Caregivers express anxiety about interrupted payments and reapplication.
  • Multiple commenters urge people to immediately download and archive SSA records (statements, payment history, earnings data) as PDFs and XML from SSA.gov.
  • Some advise keeping paper copies and any supporting documents in case of system failure or data loss.

Feasibility of a Full Rewrite in “Months”

  • Near-unanimous view that rewriting tens of millions of lines of mission‑critical code in months is impossible.
  • Experienced developers describe multi‑year rewrite efforts for far smaller systems, often with multiple failed attempts.
  • Several emphasize that a safe replacement would require years, parallel running, shadow comparisons, and gradual cutover.

COBOL, Mainframes, and “Legacy”

  • Strong pushback against treating COBOL itself as the problem; many argue the real issues are decades of accumulated business rules, tech debt, and z/OS-era design.
  • Others counter that tiny talent pools and weak ecosystem support make COBOL “legacy and bad” from a workforce and cost perspective.
  • Some highlight mainframes’ efficiency, reliability, and mature tooling, and warn that replicating batch jobs, security controls, and operational processes is far harder than “porting code.”

Staffing, Cost, and Motivations

  • The rewrite is framed by some as an ideological or ego project: proving that outsiders and “10x engineers” can do what agencies allegedly failed to do.
  • Others suspect more mundane drivers: mainframe licensing cost, COBOL hiring difficulties, or a desire to weaken Social Security by breaking its infrastructure.
  • Commenters argue COBOL specialists could be trained or paid instead of attempting a risky overhaul.

AI, Tooling, and Overconfidence

  • Some speculate the team will lean heavily on LLMs or transpilers; others respond that modern tools don’t eliminate the need to understand complex, evolving policy logic encoded over decades.
  • A few hold a minority view that better tooling might make success possible, though even they call the timelines hubristic.

Risk, Politics, and Fallout

  • Many predict large‑scale payment disruptions, litigation, and real harm to millions of retirees.
  • Several argue any failure will be politically reframed (e.g., blaming “fraud” or opponents) rather than owned as a technical misstep.