CASE STUDY
May 8, 2024

Gen AI Prompt & Output Evaluation For Search Algorithms

Gen AI Prompt & Output Evaluation For Search AlgorithmsAbstract background shapes
Table of Contents
Talk to an Expert

Redefining Search Excellence by Elevating the End-Use Experience

A Fortune 100 technology company wanted to fine-tune its search algorithm and elevate its AI-augmented search experience, delivering natural and engaging answers to users’ questions. 

They engaged Sama’s team of data experts to help evaluate model prompts and responses for alignment with pre-defined criteria including accuracy and compliance in order to help the model master the nuances of the English language. 

Every Sama model evaluation project begins with a collaborative launch session where your SMEs and ours align on expectations, requirements, and deliverables. For this project, our team of data experts would evaluate the quality of prompts and the outputs across three key areas:  

  • Factual accuracy: comparing information with reliable web sources
  • Policy compliance: ensuring responses were appropriate and free from negative impacts
  • Helpfulness and clarity: assessing a response met the user's intent in a clear, useful way

Access the full case study:

at

at
RESOURCES

Related Case Studies

No items found.