Functional Testing
Verify that features work as intended across complete user flows rather than only isolated screens.
Professional Profile
Product QA & Testing
I test products across real user flows, devices, data states, permissions, persistence, failure conditions, and release-critical edge cases.
My QA approach combines exploratory testing, structured regression, browser and physical-device validation, data-integrity checks, and evidence-based release decisions. The goal is not simply to find bugs. It is to understand what failed, isolate why it failed, verify the correction, and determine whether the product is actually ready to move forward.
01 / Capabilities
QA Core Capabilities
Exploratory Testing
Try unexpected paths, unusual inputs, repeated actions, interrupted flows, and combinations that happy-path testing can miss.
Regression Testing
Recheck previously working behavior after changes and identify when a fix creates a new failure elsewhere.
Device & Release Verification
Validate builds on physical devices, work through release gates, and separate “code completed” from “actually verified.”
Data & State Integrity
Test persistence, synchronization, stale state, migrations, destructive behavior, and whether user data survives expected product actions.
AI / Model Behavior Testing
Evaluate AI-assisted features for incorrect matches, unstable behavior, edge cases, and situations where the product needs safer fallback behavior.
02 / Approach
How I Test
I don’t treat “the build succeeded” as proof that a product works.
I test the user flow, the state behind it, what happens when things are repeated or interrupted, what breaks after a change, and whether the result is actually ready for the next stage.
03 / Case Studies
Selected QA Case Studies
Two deeper case studies showing different QA risk profiles: physical-device ML and data, permissions, and persistence.
CATch
Overview
CATch is a mobile recognition product designed to identify known street cats from photographs. Testing a recognition product involves more than validating screens and buttons: quality can depend on image capture, preprocessing, model initialization, embeddings, candidate matching, confidence handling, persistence, device performance, and the interaction between the mobile application and the underlying ML runtime.
QA scope
Investigation / readiness example
A serious physical-device crash persisted after the model-loading implementation was changed from a large memory-buffer handoff to a file-backed loading path. I reproduced the failure again rather than treating the implementation change as proof of a fix. The repeated failure showed that the original explanation was incomplete, so the release gate stayed closed while the investigation moved toward better-symbolicated physical-device crash evidence instead of additional speculative fixes.
A failed hypothesis is still useful evidence. Complex failures should narrow the investigation, not lower the release standard.
Pawfolio
Overview
Pawfolio combines persistent user data, photos, responsive interfaces, authentication architecture, permissions, and cloud-backed storage. An interface can appear correct while still failing through refresh behavior, broken persistence, permission leakage, multi-user access, storage isolation, or inconsistent behavior between devices.
QA scope
Investigation / readiness example
The QA process deliberately separated implemented behavior from behavior proven in the target environment. Local schema, role rules, permission logic, and storage policies were not treated as final proof of production security. Live readiness required real verification of authentication, session handling, data isolation, storage isolation, and cloud behavior.
A working interface is not proof that the data and permission model underneath it is safe.
04 / Additional Evidence
Additional Product Testing
A few examples beyond the two case studies above, showing how the testing surface changes across different kinds of products.
Veterinary Workflow Automation
Testing covered staff-role boundaries, request routing, case-board state changes, manager workflows, repeated handoffs, and large-case simulations across callback, results, refill, and triage scenarios.
GED Learning Product
Testing focused on complete learner flows, state transitions, regression between product stages, and readiness gates as the product moved from Founder Alpha through Beta and production-readiness work.
Directory Factory
Testing spans evidence quality, duplicate and classification checks, ZIP and service-area matching, fail-closed publication rules, sitemap coverage, and population-readiness audits.
Resetly
Testing focused on timer behavior, onboarding and preferences, notifications, app lifecycle changes, persistence, and regression across stateful mobile flows.
05 / Environment
Tools & Working Environment
Hands-on with physical iPhone testing, Xcode build and device workflows, Git/GitHub feature-branch workflows, CI and test results, issue reproduction, regression checks, structured release gates, screenshots and recordings for evidence, and AI-assisted development environments.
06 / Fit
Good Fit
- Product QA
- Manual QA
- Exploratory Testing
- Mobile App Testing
- Web App Testing
- Regression Testing
- Release Verification
- AI-Feature Testing
- Product Acceptance Testing
- Early-Stage Product QA
07 / Contact
Need another set of eyes on a product?
If you’re building something and need someone to test the product rather than simply confirm that it runs, tell me what you’re working on.