At OpenTrain AI, I authored and evaluated multi-step reasoning prompts across healthcare, logic, and scientific disciplines. I reviewed model responses for factual accuracy and scientific validity, then provided qualitative feedback and RLHF scores.
At Minddrift AI, I ranked model outputs using RLHF techniques, focusing on reducing hallucinations in technical and medical contexts. I also designed multi-turn prompts and carried out adversarial red-teaming tasks.
At Toloka, I audited prompt-response datasets for scientific accuracy and annotated complex language data to support semantic evaluation and natural language processing pipelines.
My background includes a Bachelor of Medical Laboratory Science, with a specialization in hematology, molecular science, and clinical diagnostics, and current graduate study in public health and decision science. In my current Laboratory Quality Assurance Officer role at Institutional Diagnostics & Compliance, I audit scientific records and develop standard operating procedures in line with ISO quality management standards.
