Kosisochukwu Maureen Obiekwe · Senior QA Engineer
Six years testing fintech, AI and marketplace products across web and mobile. Developers call me Human AI or QA Queen. My bug reports get fixed fast because they're impossible to misread, and I test the way real users break things: slow networks, edge cases, high traffic.
14
Critical defects caught before launch
NervaGuard
80
Defects fixed and verified before the first founders joined
FoundrHive
15
OWASP API security findings
Anvila
25
Requirement conflicts resolved before any code
Neeruam
Reconxi
GroupFund
GroupStage
FoundrHive · AI-guided startup platform
A founder on a paid plan was told they were out of AI tokens. The plan's monthly tokens had never been credited, one AI feature ignored the balance while the other enforced it, and every Buy button failed. I reported the cause as Urgent, so the team fixed it once instead of three times.
All three fixed and verified on retest.
Paid plan's monthly tokens not credited
Client details removed.
Most recent, 2026. A few products where testing changed the outcome before launch.
An AI WhatsApp operations desk for small businesses, handling bookings, stock checks and Paystack payments. Test evidence is documented for every release and shared on request.
25
Requirement conflicts resolved
5
CI quality gates
Sole QA on an AI-guided startup platform, from the proposal and test plan through to the Go release decision. Found and verified fixes for 80 defects, including billing and AI defects, before the first founders were invited.
80
Defects fixed and verified
Owned the Go/No-Go release decision.
Led independent QA on a secure document-sharing platform, covering functional, API and security testing from requirements gaps to a conditional release sign-off. Security-focused testing caught 14 Critical defects before launch.
14
Critical defects before launch
Issued the conditional release sign-off.
Sole QA engineer on a two-sided marketplace connecting African music creatives with buyers, handling real Paystack payments, escrow, legally binding contracts, and provider payouts. Tested on production with no staging environment.
189
Requirements extracted
115
Test cases
25
Bugs, 5 critical
QA Engineer on Anvila, a platform that turns plain-English descriptions into structured, GitHub-ready AI agent packages with public registry browsing and a paid private-package tier via Stripe.
224
Test cases
15
OWASP security findings
22
WCAG accessibility issues
QA Engineer testing a real-time multiplayer party game platform with 48+ browser-based game modes, WebSocket-driven room synchronization, and a zero-auth join flow. Focused on real-time state consistency, concurrent player edge cases, and cross-device gameplay.
48+
Game modes tested
128
Test cases
60
Bugs logged
Quality engineering for the internet resource-management platform behind 250,000+ registered users and 2,600+ member organisations. The registry manages roughly 112 million allocated IPv4 addresses.
Sole QA on a professional matchmaking app (React Native/NestJS, iOS and Android). Owned exploratory and API security testing through to release-readiness sign-off, and flagged critical identity-verification and payment-to-entitlement risks across mobile and admin before go-live.
Owned end-to-end QA for a peer-to-peer rental marketplace with listings across Sports, Outdoors, Tools, Electronics, Kitchen, Clothing, and Party & Events, covering web, iOS, and Android.
Artifacts under NDA
Plus 40+ more products tested since 2019.
Most of my client work is under NDA.
I can't publish client test plans or bug reports, but I'll walk you through redacted ones on a call, with client details removed, the same way the FHQA-36 ticket above is shown.
Ask for a redacted walkthroughKIND WORDS
Most of my work is under NDA, so here is what a founder I tested for said in public.
“Her bug reports are clear enough for developers to act on without follow-up questions. She marks things Blocked honestly instead of faking results. That integrity is rare.”
From Michael's public post on LinkedIn.
Hands-on security testing of web apps and APIs, run alongside functional QA.
Cross-user authorization bypass attempts via URL manipulation and direct API probing, confirming row-level security holds between accounts.
Tested on: AfroTape Marketplace
XSS payloads in bio, listing and search fields, SQL injection in search, plus stored XSS and SQL syntax handling across API inputs.
Tested on: AfroTape Marketplace · Anvila
JWT inspection for sensitive data, multi-tab session invalidation (found a zombie session after logout), and account enumeration tests on login, signup, forgot-password and resend-verification endpoints.
Tested on: AfroTape Marketplace · Anvila · SharePact
Independent assessment with 15 reproducible findings, including missing rate limiting, a 503 crash from oversized payloads, and JWT and internal UUID exposure in API responses.
Tested on: Anvila
FEATURED FRAMEWORK
One reusable quality system: written for people, wired for AI agents.
The methodology I run on every product, with or without a task brief. The Master Guide covers the full QA lifecycle, from discovery and spec review to release sign-off, with risk-based decision frameworks, execution playbooks and artifact templates. The Agent Kit turns the same method into instructions an AI agent follows, so testing can start in a real browser and keep running in CI without someone re-explaining the process every time.
8-phase lifecycle, self-directed intake checklist, decision frameworks, and playbooks for functional, API, OWASP API security, WCAG accessibility, k6 performance and regression testing.
Open folder ↗Non-functional requirements (ISO/IEC 25010), an 8-category edge-case framework, defensible findings, test data management, Gherkin conventions, and the standards behind it all (ISO/IEC/IEEE 29119, IEEE 829).
Open folder ↗An AGENTS.md rules file (never fabricate a result), 11 reusable skills from spec review to release readiness, Playwright via MCP, and a starter auth regression suite running on GitHub Actions.
Open folder ↗B.Sc. Software Development
International University of Applied Science, Germany
Open to remote QA Engineer and SDET roles. Six years across fintech, AI, gaming and marketplaces, with security and accessibility testing in the same pair of hands.
Email meA 30-day QA pilot for startups shipping payments, auth or AI. No long-term commitment.