Taking over a half-built product. Delivering against the clock.
Founding Minds joined a dental care startup midway through development, inheriting an unfinished platform, an unclear scope and evolving requirements. We brought structure to the engineering effort and delivered multiple releases against funding-linked milestones.
The situation
The company set out to become a serious player in dental care services. We joined midway through the project cycle.
Planning and execution were already underway, and the brief and scope of work were not clear. Inheriting the project meant inheriting the code and the decisions behind it, many of them undocumented. The requirements kept changing as the startup refined what it was building.
The release schedule was tied to funding, so the deadlines were external.
What made it hard
Reading someone else’s half-built system.
The codebase was a record of decisions made by the previous team, some deliberate and some accidental. Telling them apart before changing anything was the first job, and it was harder than on an older system because nothing had been in production long enough to prove itself.
Changing requirements as a permanent condition.
The company did not yet know exactly what the market wanted, so a frozen specification was not an option. The architecture and release process had to expect change, with releases short enough that a change of direction cost weeks, not quarters.
Three payment regimes in one billing flow.
HSA and FSA accounts, insurance, and inbuilt financing each carry different rules about what is eligible, who approves it, when money moves and what happens when a claim is rejected. Presenting that to a patient as one clear price is a domain modeling problem before it is an interface problem.
Image analysis on photographs taken by patients.
The system evaluates dental photos uploaded by the patient to support online assessment and treatment planning. The images arrive in uncontrolled conditions: phone cameras, domestic lighting, varying angles, no clinical framing. That is harder than working from clinical imaging, and the output is designed to support a clinician’s assessment, not replace it.
What was built
Registration for both patients and dentists, with real-time registration supported by active smart assist.
A flexible billing system spanning HSA and FSA accounts, insurance, and inbuilt financing. Treatment is offered as tiered packages, Mild, Moderate and Comprehensive, which turns a quote into a choice a patient can make.
An image analysis capability that takes patient-uploaded dental photographs and supports online evaluation and treatment planning.
Delivered on AWS with Azure DevOps, run as Agile, across multiple releases.
What changed
Delivered across multiple releases
The team brought structure to an inherited development effort and delivered product releases aligned with the startup’s funding milestones.
Following the product launch, the client raised $13 million in additional funding.
The engineering decision that mattered
Taking over an unfinished product required more than writing new features. The team had to understand the existing system, establish a workable release process, and make room for changing requirements without losing control of delivery.