Independent software engineering practice

Complex software.
Clear technical decisions.

I help businesses and engineering teams diagnose, recover and complete systems where application logic, data and infrastructure intersect.

ROOT
CAUSE
01 / ObserveExpected vs actual
02 / TraceApplication & data
03 / RepairSmallest valid change
04 / VerifyProven outcome
First Class HonoursB.Sc. Special in Computer Science
3.82 / 4.00GPA · Gold Medal
Business systemsPayroll · PMS · POS · Access
End-to-endInterface · API · Data · Deployment
01 / Where I am useful

When another surface-level patch is not enough.

Defined technical intervention for an existing system, failure or delivery risk—not generic development sold by the hour.

Delivery recovery

An unfinished or abandoned application

Establish what really works, identify blocking gaps and turn the remaining work into a defensible delivery sequence.

Production investigation

A defect that keeps returning

Reproduce the failure and trace it across application, permission, integration and data boundaries before changing code.

Infrastructure recovery

A release that will not stay healthy

Work across Linux, Nginx, Docker, PM2, SSL, DNS and database connectivity to restore a repeatable baseline.

Launch assurance

A codebase approaching real customers

Review architecture, authorization, data handling, failure modes and operational readiness, ranked by business risk.

02 / Evidence

Experience where business rules meet production reality.

These are systems and operational domains I have worked on—not speculative capabilities or invented performance claims.

CASE CONTEXT / 01

Attendance, payroll and biometric data

Vue.jsCodeIgniterBusiness rules

Implemented and supported leave, overtime, breaks, short-time and configurable payroll calculations while handling device connectivity and inconsistent attendance records.

Engineering concern

Rules interact across shifts, time boundaries, approvals and client-specific configuration.

Relevant strength

Following a defect from recorded event through calculation, storage and displayed result.

CASE CONTEXT / 02

Hospitality operations

NestJSReactLaravel

Worked across reservations, room and rate management, billing, POS, housekeeping, OTA integration, folio ledgers, guest payments, cashiering and night-audit workflows.

Engineering concern

Operational events must remain consistent across money, inventory, rooms, guests and reporting periods.

Relevant strength

Understanding workflows as connected business transactions rather than isolated screens.

CASE CONTEXT / 03

Multi-tenant business applications

JWTRBAC / ABACMySQL / PostgreSQL

Built tenant-scoped access, authentication, user management and permission-aware APIs and interfaces for applications serving multiple organizations.

Engineering concern

Every query, operation and interface decision must respect organizational and authorization boundaries.

Relevant strength

Working across API, data model and frontend enforcement instead of treating access control as middleware alone.

03 / Diagnostic method

Make the problem smaller before making the change larger.

The goal is not only to produce a patch. It is to leave the failure, decision and remaining risk understandable.

01 / FRAME

Establish the boundary

Expected behaviour, observed failure, environment and business impact.

02 / PROVE

Reproduce before changing

Gather evidence and isolate the smallest defensible root cause.

03 / REPAIR

Change the right layer

Fix the cause without silently expanding scope or hiding adjacent risk.

04 / VERIFY

Test the outcome

Confirm the success condition and document what remains.

04 / Engagement

Small enough to start safely. Specific enough to finish.

Scope, access, price and success conditions are agreed before implementation begins.

Start with the actual problem.

Send the symptoms, stack, urgency and any error evidence you already have. I will tell you whether it fits my experience and what I need to assess next.