I translate breakdowns in clinical workflows into specs engineering teams can build from. That means mapping how information actually moves through a clinical or operational process, finding where it stalls or gets lost, and turning that into a model a backend team can implement directly.

The Clinical Workflow Signal Audit is the clearest example: an ICU signal-to-action latency model with an escalation state machine and risk-tiered routing logic, built to show where alerts break down before they reach a clinician. The Kenya Health Facilities Dashboard applies the same approach at a system level, a public API covering 10,483 facilities across all 47 counties, structured to surface gaps like the facility density difference between counties such as Mandera and Nyeri.

My background is in biomedical engineering, with operational experience inside Kenya's Level 5 and Level 6 public referral hospitals, the tier that handles the highest patient volume and the most complex cases. That combination is why I read clinical workflows the way an engineer reads a system: as a set of states, transitions, and failure points, not as a general process to describe in the abstract.

I work with founders and Heads of Product at early-stage health-tech companies who need someone who can sit between clinical reality and software architecture, and hand back something the engineering team can actually build.