For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Nelson John Kuma

Nelson John Kuma

LLM Evaluation and Text Generation Specialist in French and English

Cameroon flagYaounde, Cameroon
$10.00/hrEntry LevelOther

Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

AudioAudio
DocumentDocument
ImageImage

Top Task Types

Classification
Question Answering
RLHF
Text Summarization
Translation Localization

Freelancer Overview

I am a results-oriented professional and passionate AI Trainer, having worked as a French AI Contributor with Outlier to build the world's most advanced generative AI model. In this position I executed the following tasks: - I trained AI generative models by initiating queries that were used by these models to generate AI responses. - I rated each of these AI responses based on some predefined rating criteria. - I selected the ideal response for further processing and editing. - I supported my choice with a brief justification of why I described a response as being better than another.

Entry LevelFrenchEnglish

Labeling Experience

Language Preference Writing and Rewrite - RLHF

OtherTextRLHFEvaluation Rating
This project consisted in initiating queries based on specific domains, and evaluating the two responses generated by the two models. Some tasks included 2 to 4 turns, whereby the contributor had to initiate the number of queries that were required and respect the domains and ensure a logical follow up of the queries. The chosen responses were eventually rewritten and justification were provided to support the judgement that was made.

This project consisted in initiating queries based on specific domains, and evaluating the two responses generated by the two models. Some tasks included 2 to 4 turns, whereby the contributor had to initiate the number of queries that were required and respect the domains and ensure a logical follow up of the queries. The chosen responses were eventually rewritten and justification were provided to support the judgement that was made.

2024

Cocoa Factuality

OtherTextQuestion AnsweringEvaluation Rating
The project consisted in initiating queries on a specific theme and evaluating the factuality of the AI responses based on some criteria. A detailed justification was later on provided to back up the claims of the chosen AI response.

The project consisted in initiating queries on a specific theme and evaluating the factuality of the AI responses based on some criteria. A detailed justification was later on provided to back up the claims of the chosen AI response.

2024

Cypher - Evals

OtherTextEvaluation Rating
The project consisted in evaluating the responses of two AI models according to a set of criteria. The prompts were generated as well as the responses. The best AI response was selected and the contributor had to back his choice with a detailed justification. Also, there was a possibility of rewriting the selected response.

The project consisted in evaluating the responses of two AI models according to a set of criteria. The prompts were generated as well as the responses. The best AI response was selected and the contributor had to back his choice with a detailed justification. Also, there was a possibility of rewriting the selected response.

2024

Cypher - RLHF

OtherTextQuestion AnsweringText Summarization
The project consisted in training two AI models by initiating queries and evaluating model responses based on specific criteria. The goal of this project was to reinforce the model's interactive abilities on different subjects and improve its robustness.

The project consisted in training two AI models by initiating queries and evaluating model responses based on specific criteria. The goal of this project was to reinforce the model's interactive abilities on different subjects and improve its robustness.

2024

Winter Wonderland - RLHF

OtherTextRLHFEvaluation Rating
This project consisted in initiating queries based on specific domains, and evaluating the two responses generated by the two models. Some tasks included 2 to 4 turns, whereby the contributor had to initiate the number of queries that were required and respect the domains and ensure a logical follow up of the queries. The chosen responses were eventually rewritten and justification were provided to support the judgement that was made.

This project consisted in initiating queries based on specific domains, and evaluating the two responses generated by the two models. Some tasks included 2 to 4 turns, whereby the contributor had to initiate the number of queries that were required and respect the domains and ensure a logical follow up of the queries. The chosen responses were eventually rewritten and justification were provided to support the judgement that was made.

2024 - 2024

Education

U

University of Dschang

Bachelor of Science in Management, Economics and Management Sciences

Bachelor of Science in Management
2019 - 2022
G

Government Bilingual Practicing High School Yaounde

General Certificate of Education (GCE) Advanced Level, Economics, History, Geography, Information and Communication Technology

General Certificate of Education (GCE) Advanced Level
2017 - 2019

Work History

M

MTN Cameroon

Business Sales Agent

Yaounde
2024 - Present
O

Outlier

Remote French AI Trainer

Yaounde
2024 - Present