Hangtime connects people through shared hobbies — but the story worth telling here isn't the product concept, it's what it takes to keep a real system running for a solo founder over years. When Shees was first engaged, the codebase's own history was already a problem: it existed across more than one repository, and the version handed over was significantly out of date. Tracking down the actual current codebase meant going back through a previous developer no longer actively involved.

Recovering and consolidating the real system — before any feature work could happen, the actual current codebase had to be located and confirmed, treating the handoff as an investigation, not a starting point.

Diagnosing and repairing recurring, externally-triggered failures — on three separate, unconnected occasions over roughly two years, push notifications stopped working because Google/Firebase deprecated the API the feature depended on, not because of anything built or changed on either side. Each occurrence was diagnosed and restored without an extended outage or a full rebuild.

Untangling a scope dispute before it became a trust problem — the founder believed a shipped milestone was incomplete because notifications weren't working; the notification failure was the deprecated-API issue above, structurally unrelated to the feature that had, in fact, shipped. Resolved through a direct conversation about what was actually in scope.

Catching the gap between "works between us" and "works" — multiple times, a feature tested clean between founder and developer, then failed the moment a real user touched it: a contact list large enough to change fetch behavior, a device-specific permission-flow edge case, legacy account data that didn't match new logic's expectations.

Adapting to a real, fluctuating budget — the relationship ran across 7+ separate fixed-price milestones over multiple years. At one point the founder's available budget dropped sharply; scope was deliberately narrowed to what mattered most, then work resumed at fuller scope once budget allowed again.

Most portfolio stories are about what got built. This one is about what it takes to keep something built by someone else — or built years ago — actually working: recovering institutional knowledge nobody wrote down, absorbing failures that have nothing to do with your own code, and treating a scope disagreement as a conversation to have directly rather than a fight to win.