Priya Ramanathan

Analyze A/B test results with detailed reasoning.

Systematically detect and quantify bias in model outputs.

Trace hallucinations to their underlying causes systematically.

Compare models through systematic reasoning steps.

Calibrate output length expectations using examples.

Score ethical compliance using example-based guidelines.

Evaluate response coherence using example ratings.

Rank models using pairwise comparison examples.

Score instruction complexity based on example difficulties.

Verify factual grounding using source-based examples.

Score prompt effectiveness using example-based calibration.

Validate API responses using example schemas.