Case Study: Fortune 100 Technology Company achieves higher search accuracy, relevance, and user satisfaction with Sama's Gen AI prompt & output evaluation

A top 10 technology company turned to Sama to evaluate clarity, helpfulness, and compliance in AI prompts and outputs

A Fortune 100 Technology Company turned to Sama to fine-tune its search algorithm and elevate its AI‑augmented search experience by evaluating the clarity, helpfulness, and compliance of model prompts and outputs. The customer engaged Sama’s model evaluation and human‑in‑the‑loop (HITL) data‑expert services to align on expectations and assess prompts and responses against criteria like factual accuracy, policy compliance, and helpfulness.

Sama’s team ran collaborative launch sessions, reviewed outputs for accuracy and safety, and rewrote inaccurate prompts and responses to create additional training datasets for supervised fine‑tuning. The work increased model confidence and delivered measurable gains—significant improvements in search accuracy, relevance, and user satisfaction—while helping accelerate model deployment (Sama cites up to 3x faster time to production).