Statement of Work Atlantis/Avalon SECTION I: Scope of Work

The Atlantis/Avalon project involves evaluating search queries and results to provide relevance ratings, aimed at training AI to enhance client services. Participants must pass certifications and be native speakers or have significant cultural understanding of the language they work in. Comments on tasks are mandatory and must be unique and specific, with strict penalties for non-compliance.

Uploaded by

yoyo

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

57 views1 page

Statement of Work Atlantis/Avalon SECTION I: Scope of Work

Uploaded by

yoyo

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 1

Statement of Work

Atlantis/Avalon
SECTION I: Scope of Work

The Atlantis/Avalon project consists of different evaluations/workflows that involve being

provided a search query and a result, and then providing a relevance rating for the result based
on the specific guidelines for the evaluation. The different lines of business for the project are
apps, music, podcasts, but are subject to change depending on client need.

The goal of the Atlantis/Avalon project is to train Artificial Intelligence using Machine Learning to
improve the client’s services.

The information produced from these tasks will be used for evaluation purposes only.

As there are multiple evaluations within the Atlantis/Avalon project, you will be assigned
evaluations based on necessity, meaning we can only offer you access to evaluations with a
current need. However, once you are in production, you should expect to be offered access to
other evaluations whenever the necessity arises.

To start the collaboration on the Atlantis/Avalon project, it is compulsory to read and understand
the Guidelines then pass the project certifications available in OneForma. You must be a native
speaker of the language you will be working on (or at least have lived minimum 5 years in country
and have a good cultural understanding) and must reside in country for the applied locale.

SECTION II: Description of Tool and Rating Interface

You will be doing the rating work in the client’s tool. You will receive access to this tool after you
clear OneForma´s certification process.

Once in the tool, you will have access to the task guidelines, evaluation-specific qualification,
and you will gain access to the production tasks once you pass.

Once in production, please note that comments are ALWAYS MANDATORY and tasks that are
missing comments will be considered incomplete tasks and therefore cannot be paid for.
In addition, comments must be specific, accurate, unique, and personally crafted. Any user
caught copy-pasting comments, using same comments as other raters, gibberish, random or
one-letter or character comments will be flagged and will be at risk for removal and non-
payment. It is also not allowed to include AI generated comments (ChatGPT) and comments
that are gibberish or random letters and characters.

SECTION III: Research Expectation

We encourage you to do some research for every query, as this will help you understand the
user’s query and intent. Here’s below the research expectation for every rating:

www.centific.com