Kurmet A.

AI / LLM Engineer, Applied Machine Learning, AI Systems Evaluation

I am an experienced Senior AI / LLM engineer and applied machine learning practitioner with 20+ years of engineering and analytical experience. I work with real-world LLM systems, focusing on how models actually behave in practice — under ambiguity, pressure, and adversarial input — not how they are described in documentation. Core focus Applied LLM engineering & evaluation black-box testing of LLM systems stress-testing for hallucinations, authority leakage, and overconfidence diagnostic vs prescriptive behavior analysis ambiguity handling and failure mode detection Machine learning and data-driven analysis Large-scale data processing, validation, and reliability I evaluate systems by observed behavior, not by stated intent. Claims such as “non-prescriptive”, “diagnostic-only”, or “non-authoritative” are treated as hypotheses to be tested, not assumptions. Background that matters My background includes years of building production systems, analytical models, and financial forecasts used in high-stakes environments. I have operated at executive and government level, where correctness, restraint, and accountability are mandatory. This experience allows me to reason rigorously, detect subtle failure modes, and remain disciplined under uncertainty.