Artificial Intelligence QA Manager

Job Description A QA Engineer for AI Initiatives is responsible for ensuring the quality, reliability, fairness, and performance of AI/ML-powered products and systems. Unlike traditional QA, this role requires deep understanding of non-deterministic model behavior, data quality, and AI-specific failure modes such as hallucinations, bias, and model drift. Key Responsibilities - Design and execute test strategies specifically for AI/ML models, LLM-based applications, and data pipelines - Develop automated test frameworks for model validation, regression testing, and performance benchmarking - Evaluate model outputs for accuracy, consistency, relevance, hallucination, and bias across diverse inputs - Test RAG (Retrieval-Augmented Generation) pipelines, chatbots, recommendation systems, and other AI-driven features - Collaborate with data scientists and ML engineers to define acceptance criteria and quality thresholds - Build and maintain evaluation datasets, ground truth sets, and adversarial test cases - Monitor models in production for drift, degradation, and anomalous behavior - Validate data quality, data pipelines, and feature stores that feed AI systems - Document defects, edge cases, and failure patterns specific to AI behavior - Ensure AI systems meet ethical, fairness, and compliance standards (bias audits, explainability checks) Required Skills & Qualifications - Bachelor's or Master's degree in Computer Science, Engineering, or a related field - 3–6 years of QA experience, with at least 1–2 years in AI/ML quality assurance - Strong proficiency in Python for test automation and data analysis - Familiarity with LLM evaluation frameworks (e.g., RAGAS, DeepEval, Promptfoo, LangSmith) - Hands-on experience with testing tools: Pytest, Selenium, Postman, or similar - Understanding of ML lifecycle — training, validation, deployment, and monitoring - Knowledge of data quality tools and pipeline testing (Great Expectations, dbt tests) Nice to Have - Experience with prompt engineering and red-teaming LLMs - Familiarity with MLOps platforms (MLflow, SageMaker, Vertex AI) - Knowledge of vector databases and embedding quality evaluation - Understanding of AI safety, responsible AI principles, and fairness frameworks - Experience with A/B testing and shadow deployment strategies Soft Skills - Analytical and inquisitive mindset — comfortable challenging model outputs - Ability to think like both a user and an adversary (red-team thinking) - Strong documentation and communication skills - Collaborative approach with data science, engineering, and product teams - High attention to detail with a quality-first attitude Preferred Location: Bangalore