JJ Maxwell was diagnosed with squamous cell carcinoma in his mouth in May 2026. Instead of writing a memoir, he wrote a checklist. The result reads like a bug report filed against the healthcare system by someone who understands that systems don't optimize for you unless you force them to. The core insight isn't medical — it's operational. Maxwell's first lesson is pure scheduling arbitrage: call back, ask for cancellations, compress timelines. His wife pushed him to call back on a 4-week ENT wait; he got in that week. In cancer, the delta between four weeks and four days is not a convenience difference. It's a survival variable. The AI warning is quietly damning. Maxwell reports that multiple models pushed him toward "you don't have cancer" before diagnosis. This isn't an indictment of AI in medicine broadly — it's a specific, documented failure mode where probabilistic reassurance collides with a patient who needs urgency. The lesson: AI is a research tool, not a triage nurse. The genetic testing delay is the piece's sharpest edge. Maxwell's surgeon said he'd order comprehensive testing, then didn't, then referred him to an oncologist, causing a multi-month delay because results were eventually run on the surgery tumor rather than the biopsy tumor. Nobody flagged the gap. The system didn't fail catastrophically — it failed through accumulated small neglects, each one someone else's problem. Maxwell's practical toolkit is genuinely useful: NCCN guidelines (free with a non-professional account), the Sijbrandij Foundation's resources (Sid Sijbrandij, GitLab cofounder, built this after standard of care ran out for him), dictating post-appointment notes into AI-generated summaries, and the blunt advice to get on an airplane for a second opinion at a top cancer center. The Hacker News community added a critical addendum: if you're within 2-3 hours of a major cancer center, go there. The advantage isn't better doctors — it's speed and institutional density. At MD Anderson, Maxwell got a pathology question answered in 20 minutes because the doctor called the pathologist by first name. At a community hospital, that question dies in a queue. The piece's most underappreciated recommendation: talk to someone who has already had your surgery. Maxwell didn't, and regrets it. Surgeons describe procedures; patients describe experiences. The information asymmetry between "this is a free flap reconstruction" and "you will not be able to swallow for three weeks" is enormous. This is not a medical guide. It's an operations manual for being your own project manager inside a system that will not project-manage for you. The implicit argument is that cancer treatment in the US is a system with capable components and terrible integration, and the patient is the only person with both the incentive and the access to do the integration work.
