AI Linguistic Trainer

Role Title: AI Linguistic Trainer

Overview:
We have several backend LLM / AI Training processes where we either evaluate or label LLM
data based on the given guidelines, common sense and a bit of online research. By LLM data,
we mean prompts (that users give to LLMs) or responses (that LLMs provide back to users).
This data can be in the form of text, image, audio & video.

Note: This data can have explicit / sensitive content too.

Key Responsibilities:
We are seeking skilled individuals to help us with the above type of tasks / processes. In this
role, you will be required to label / evaluate the LLM data using the guidelines provided, your
common sense and a bit of online research. Below are some of the parameters that you will
have to evaluate.

1. Safety / Harmlessness: Does the data have any dangerous / sexually explicit /
discriminatory, etc. type of content or not?

2. Text Content Writing Style
3. Factuality / Truthfulness: Are the claims given in the content factually correct?

4. Verbosity
5. English Grammar
6. Instructions Adherence

7. Relevance

In this role, you will be frequently required to:
1. Self learn and train yourselves based on the given set of guidelines
2. Use your general knowledge, common sense & judgment to evaluate / label the data
3. Researching using online resources to validate facts and their accuracy
4. Check english grammar

Must have skills:
● Excellent verbal, reading and writing English skills.
● Excellent reasoning skills.
● Ability to do an online research to validate facts from unfamiliar domains.
● Ability to learn new tools with ease.
● Ability to work with text/audio/image/video base explicit content (dangerous / hate
speech / sexual / discrimination).
● Ability to work with all types of data types - text, audio, image, video.

Experience:
● Prior experience in working with safety content evaluations and data labeling is a plus.
Location: Hyderabad / In-Office