**Evaluation & Proof of Value** — Design and guide customer evaluation frameworks (evals) to measure AI performance for specific use cases. Create compelling demos, reference architectures, and lightweight POCs that demonstrate the art of the possible. Agree acceptance criteria for quality, latency, cost and reliability, linking evaluation results to customer business goals and a decision on production deployment
Proficiency in programming languages commonly used in AI (such as Python, C/C++, or Java) and relevant ML frameworks (for example, TensorFlow, PyTorch, or similar).
Design experiments and evaluate outcomes. Conduct offline evaluations, backtesting and online A/B experiments to assess model performance and business impact. Investigate data quality, feature effectiveness and evaluation bias, and use findings to guide further development.