Freelance AI Content Evaluator & Specialist (Remote Contract)
Performed side-by-side (SxS) evaluation of LLM outputs to select the optimal response based on safety, accuracy, and strict prompt compliance. Identified and documented factual hallucinations and scientific inaccuracies with concise, well-reasoned justifications and ensured adherence to administrative and linguistic constraints (e.g., word limits and character counts). Audited creative and stylistic writing prompt outputs by validating structural integrity such as poetic meter, rhyming structure, and thematic consistency. • Side-by-side model response ranking for best-fit selection • Hallucination detection and ground-truth-style justification writing • Constraint verification for formatting, sentence limits, and stylistic rules • Creative-structure validation for poetic meter, rhyme, and themes