Ismail Isleem

• Build AI-powered test-data generation systems using LLMs and dual-agent validation workflows to produce deterministic positive and negative datasets with structured quality feedback.
• Build internal Python QA tools, reusable libraries, and CLI utilities consumed by automation engineers across the organization.
• Design internal QA service APIs for test-data generation and dataset retrieval, with clean DTO/POJO contracts consumed by web and mobile automation pipelines.
• Engineer LLM evaluation pipelines covering prompt testing, hallucination detection, RAG faithfulness, semantic similarity, and LLM-as-a-judge scoring.
• Architect open-source automation frameworks for enterprise clients across web, mobile, API, and desktop, with reusable reporting, CI integration, and maintainable design patterns.
• Lead in-house automation tools and QA platforms, guiding architecture, sprint planning, code reviews, mentoring, and release readiness.
• Conduct code audits for internal teams and client engagements, reviewing framework design, design-pattern usage, maintainability, and test reliability.

• Built the Quality Engineering function inside the AI/NLP department, defining testing strategy, frameworks, and tooling for Arabic language models and conversational AI.
• Designed and built a full-stack ML model evaluation platform with a C#/.NET backend, service-oriented APIs, and a no-code frontend used by manual QA engineers.
• Architected an NLP evaluation engine for embeddings, POS tagging, NER, classification, categorization, and normalization, with Jenkins-based parallel execution and model-quality metrics.
• Engineered web scraping and data-processing pipelines to generate internal training datasets and reduce reliance on third-party data sources.
• Built a rule-based Arabic text normalization and categorization engine used as a preprocessing layer for NLP models and as a standalone data-science service.
• Extended the evaluation platform to test fine-tuned chatbots and conversational agents, including a production banking chatbot.
• Owned code review for the shared automation repository, approved production merges, trained QA engineers, and mentored manual testers on model evaluation workflows.

• Owned QA coverage for a US and Canada vehicle-auction platform across manual testing, web automation, mobile automation, and release validation.
• Rebuilt automation infrastructure across web, mobile, and JavaScript-based testing, applying reusable design patterns and maintainable test structure.
• Migrated the web suite from Cypress to Playwright with Python, improving stability, execution speed, and maintainability.
• Built an in-house testing lab with dedicated servers and physical mobile devices, reducing cloud-testing dependency while improving reliability and control.

• Owned QA for a logistics platform covering warehouse operations, slot management, inventory workflows, and inter-service REST integrations.
• Designed API automation in two layers: lightweight Postman collections for fast feedback and a Java RestAssured framework for functional, regression, and smoke coverage.
• Architected a cross-platform Appium framework with Java, Maven, and TestNG for unified Android and iOS execution.

• Owned end-to-end QA for a high-traffic classifieds platform, covering core web flows, user journeys, regression cycles, and release reporting.
• Built the company's first Selenium + Java web automation framework, establishing reusable components and structured regression suites.

• Tested a US-based e-commerce platform and custom ERP covering warehouse management, inventory, and DHL shipping integrations.
• Introduced the company's first Selenium + Java automation initiative for regression-heavy flows while owning manual test cases and release regression cycles in Mantis.

Local-first AI test-data tool that generates schema-valid positive and negative datasets through a dual-agent workflow, with validation feedback designed for repeatable QA use.

Suite of reusable automation frameworks for web, mobile, and API testing, powered by a shared Python core for reporting, artifacts, events, helpers, and consistent execution patterns.

Cross-platform E2E orchestrator for running API, web, and mobile automation as one end-to-end journey, with shared scenario state, CLI execution, and unified reporting.