QA Engineer · Abu Dhabi, UAE

Mohamed Elaadly

I test the systems people trust with their money — and the AI systems being trusted with everything else. Currently leading QA for a merchant acquiring platform: CIT/MIT-compliant payment flows, dynamic currency conversion, and card/settlement data integrity. Previously validated AI models for static analysis, requirements engineering, and inbox automation.

ISTQB CTFL ISTQB CTAL-TAE ISTQB CT-PT ISTQB CT-AuT
▶ Watch my videos View experience Download CV Contact me
60→87% Test coverage growth at Botim Astra Tech
3 AI systems tested end-to-end
4 ISTQB certifications
3 Junior engineers mentored

02 · About

Quality engineering, applied where mistakes are expensive.

I'm an ISTQB-certified QA engineer (Foundation, Test Automation Engineer, Performance Testing, Automotive) with hands-on experience across manual and automated testing for web, mobile, and API applications. I've worked in Agile teams alongside European clients, built Python tooling to generate test cases and reports, and moved from safety-critical automotive systems into fintech and AI-system validation — domains where a missed edge case has real consequences.

  • Now: QA Engineer, Botim Astra Tech — merchant acquiring
  • Before: AI Software Testing Engineer, SEITech Solutions GmbH
  • Based in: Abu Dhabi, United Arab Emirates
  • Languages: Arabic (native), English (C1)

03 · Experience

Where I've applied it.

Dec 2025 — Present Abu Dhabi, UAE

Quality Assurance Engineer · Botim Astra Tech

Merchant Acquiring & Payment Core — Botim Payment Platform

Leading QA for a merchant acquiring & payment core platform processing millions of transactions — CIT/MIT-compliant payment flows (the Visa/Mastercard stored-credential framework), recurring/tokenized payments, dynamic currency conversion (DCC), card data integrity, and the merchant-facing admin portal.

  • Built UI automation with Playwright and mobile validation with Appium (iOS/Android), using Java + TestNG for execution control and Cucumber for BDD scenarios.
  • Test the merchant-facing admin portal (DCC/rebate detail views, statement and report downloads, merchant auth configuration) alongside backend APIs, databases, and card BIN data-integrity pipelines.
  • Ran extensive API, database, and reporting-layer testing via REST Assured and Postman, covering DCC rebate/markup calculation, multi-currency settlement, and statement/reporting system performance against MongoDB and MySQL.
  • Identified production-impacting defects including a SQL-injection input-validation gap in a card-data pipeline, multiple DCC rebate-calculation edge cases, and data-integrity issues in automated reporting pipelines.
  • Integrated automated suites into Jenkins CI/CD with Git-based workflows; combined cross-browser frontend, portal, and backend API verification in a single pipeline.
  • Mentored three junior engineers on framework design and testing best practices — raised test coverage from 60% to 87%.
PlaywrightAppiumJava / TestNGCucumber REST AssuredPostmanJenkins CI/CDMongoDBMySQL
Jan 2025 — Dec 2025 Stuttgart, Germany

AI Software Testing Engineer · SEITech Solutions GmbH

Validated three distinct AI systems — a static-analysis model, an NLP requirements checker, and an email-classification pipeline — each requiring a different definition of "correct."

  • Tested an AI-powered MISRA C/C++ static analysis tool; built multi-size C test files with mixed violations and evaluated precision/recall across rule groups 8.x, 10.x, 11.x, 15.x, 20.x, and 21.x.
  • Validated an NLP-based requirements consistency checker analyzing SYS.2/SWE.1 requirements against ASPICE Base Practices BP1–BP4, checking feasibility, verifiability, atomicity, and consistency; surfaced logic gaps and false positives.
  • Verified "Ocean," an AI email classifier with live Gmail access, covering classification accuracy, attachment validation, corrupted-file handling, and multilingual input.
  • Ran functional, performance, security, and maintainability testing — including malicious-input and large-file stress cases — and produced test reports and bug logs to improve model reasoning.
AI/ML validationNLP testingPrecision/recall analysis Security testingExploratory testingPython

04 · AI Testing Deep Dive

Testing systems that don't have a fixed right answer.

Traditional QA checks output against a spec. Testing AI systems means testing a distribution of plausible outputs — these three projects each demanded a different strategy.

Static Analysis

MISRA-C Violations Checker

An AI tool flagging MISRA C/C++ rule violations. I built graded test files spanning multiple sizes and violation densities to measure the model's precision and recall, then verified coverage across the key rule groups (8.x, 10.x, 11.x, 15.x, 20.x, 21.x) before reporting reasoning gaps back to the team.

  • Precision / recall benchmarking
  • Rule-group coverage mapping
  • Model reasoning feedback loop
NLP / Requirements Engineering

Requirements Consistency Checker

An NLP tool assessing SYS.2/SWE.1 requirements against ASPICE Base Practices BP1–BP4. I validated outputs for feasibility, verifiability, atomicity, and consistency, then stress-tested it with malicious input and oversized files to check where the model's judgment broke down.

  • ASPICE BP1–BP4 compliance checks
  • False-positive / logic-gap detection
  • Security & stress testing
Applied AI / Automation

Ocean — AI Email Classifier

A Gmail-connected classifier sorting inbound mail into Inquiry, Order, Complaint, and Other. I verified classification accuracy alongside an automated Excel-extraction engine, testing structured-data parsing, missing-field detection, corrupted files, and multilingual inputs.

  • Classification accuracy testing
  • Attachment & corrupted-file handling
  • Multilingual data validation

05 · Open-Source QA Projects

Code you can actually run.

Nine self-contained repos demonstrating manual and automated testing across the layers a payments and AI/ML QA engineer touches daily — UI, mobile, API, performance, security, ML model evaluation, LLM testing, and structured manual test design.

UI Automation

qa-playwright-ui

TypeScript + Playwright, Page Object Model. 38 tests across login, inventory, cart, and checkout — including cross-browser CI and independently-recomputed order-total math checks.

  • Playwright · TypeScript · GitHub Actions CI
  • Positive + negative test design
  • storageState auth reuse pattern
View on GitHub →
Mobile Automation

qa-appium-mobile

Java + Appium + TestNG, Page Object Model for native Android. 30 tests across login, catalog, cart, and drawer navigation, with an emulator-based CI workflow.

  • Appium Java Client · UiAutomator2
  • Explicit waits, zero Thread.sleep
  • Android emulator GitHub Actions CI
View on GitHub →
API Automation

qa-api-restassured

Dual-track API testing: REST Assured + TestNG (Java) and a Postman/Newman collection, ~30 tests including JSON schema validation and data-driven negative cases.

  • REST Assured · Postman · Newman
  • JSON schema contract validation
  • Dual CI jobs (Maven + Newman)
View on GitHub →
Manual Test Design

qa-manual-testing

A full manual-QA artifact set: ISTQB test-design techniques, 160+ test cases, bug reports, and a requirements traceability matrix — spanning checkout, and API/DB/UI/performance/security suites for a fictional payment & wallet platform.

  • Equivalence partitioning, BVA, decision tables
  • Narrative Markdown + Zephyr-style CSV exports
  • Payment, merchant, and wallet (P2P/bill-pay) scope
View on GitHub →
Payment Gateway Testing

qa-payment-gateway-stripe

REST Assured + TestNG and Postman/Newman against Stripe's real test-mode API — 30 tests covering payment intents, refunds, decline scenarios, idempotency, and the 3D Secure automation boundary.

  • Real sandbox, not a fictional mock
  • Idempotency-key and decline-code coverage
  • Dual REST Assured + Postman tracks
View on GitHub →
Performance Testing

qa-performance-testing

k6 + JMeter load testing against a self-contained mock payment service — baseline, spike, soak, and a concurrent-refund race-condition check. Every number in the README is a real local-run result.

  • Backs ISTQB CT-PT certification with real proof
  • p95 latency, error-rate, and degradation gates
  • Verified race-safe refund handling
View on GitHub →
Security Testing

qa-payment-security-testing

OWASP API Security Top 10 applied to a payment API. An intentionally vulnerable mock service plus a dual Python/pytest + Postman suite that independently confirms 5 real, exploitable findings.

  • BOLA/IDOR, broken auth, data exposure, injection, forged webhooks
  • Professional findings report + before/after fixes
  • Every finding independently verified, not simulated
View on GitHub →
ML Model Evaluation

qa-ml-model-evaluation-testing

Precision/recall/F1 benchmarking of a real Hugging Face sentiment model across 74 hand-curated cases, plus adversarial/robustness and directional fairness-probe testing.

  • 81.8% overall accuracy — real measured result
  • Sarcasm accuracy drops to 25% (honest finding, not hidden)
  • 70% robustness under typo/emoji perturbation
View on GitHub →
LLM / GenAI Testing

qa-llm-testing-framework

Consistency, structured-output validation, prompt-injection resistance, hallucination spot-checks, and refusal-boundary testing against a deterministic mock LLM with real, seeded findings.

  • 5 findings independently verified via actual test runs
  • 10/10 prompt-injection payloads bypass a naive system prompt
  • Pluggable to real Anthropic/OpenAI backends
View on GitHub →

06 · Skills

Toolkit.

Automation & Frameworks

PlaywrightSeleniumAppiumTestNG Cucumber (BDD)PytestRobot Framework

API & Backend Testing

REST AssuredPostmanSQLMongoDB MySQLJMeter

Programming

JavaPythonJavaScriptCC++

CI/CD & Tooling

JenkinsGitHubGitLabBitbucket JiraConfluenceGrafanaBamboo

AI-System Testing

Precision / recall analysisNLP output validation Model stress testingExploratory AI testing

07 · Certifications

ISTQB® credentials.

CTFLCertified Tester Foundation Level
CT-AuTAutomotive Software Tester
CTAL-TAETest Automation Engineer
CT-PTPerformance Testing

08 · Contact

Let's talk about your next release.

Open to QA / SDET roles in fintech, mobile, and AI-product testing.

mohamedbasem512@gmail.com LinkedIn +971 50 334 4217