Our client, a leader within the entertainment space, is looking for a GenAI SDET to join their team in Orlando, FL. In this role, you will design, build, and maintain automated testing frameworks for both backend and frontend components for Gen AI platforms and technologies.
THIS IS A 22 MONTH W2 CONTRACT
Responsibilities
- Built and maintained automated testing frameworks for both backend and frontend components of Gen AI platforms, ensuring robust and scalable quality coverage.
- Developed and executed test strategies to validate AI models, agents, and knowledge base integrations, including testing for hallucinations and factual accuracy using tools like Arize.
- Designed and implemented AI guardrails to ensure safe model behavior through adversarial testing, red teaming, and ongoing research into AI safety best practices.
- Collaborated cross-functionally with developers, product managers, QA, and performance engineering teams to refine requirements and conduct system integration and load testing.
- Led the creation of test plans, documented defects, and participated in code reviews to improve testability, debugging complex issues in varied environments.
- Automated testing processes and defined new metrics for Gen AI applications to enhance continuous delivery, reduce defects, and promote responsible AI deployment.
Qualifications
- Proven ability to design, develop, and maintain test code, tools, and automation frameworks for validating Windows, mobile, and AI-based applications. Experience creating unit tests, test harnesses, and custom automation solutions.
- Demonstrated experience as an SDET or AI/ML engineer, including testing AI models and machine learning systems. Familiarity with AI safety practices such as hallucination detection, red teaming, and responsible AI evaluation.
- Proficient in Python, JavaScript, and Node.js, with the ability to translate technical requirements into feature code and automated tests.
- Experienced in RESTful API testing (e.g., Postman), database testing using SQL, BDD frameworks (e.g., Gherkin), and test plan design. Comfortable generating documentation and supporting team-wide tool adoption.