For employers

Hire this AI Trainer

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

LoginCreate account
Invite to Job
Gabriel Molina

Gabriel Molina

LLM Evaluation and Text Generation Specialist in English & Spanish.

Chile flagSantiago, Chile
$35.00/hrEntry LevelInternal Proprietary Tooling

Key Skills

Software

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

TextText

Top Task Types

Evaluation Rating
Fine Tuning
Text Generation
Text Summarization
Translation Localization

Freelancer Overview

An expert in academic and profesional text analysis, with recent contributions to Cypher Evals and Cypher RHLF projects, where I design prompts and evaluate model responses. My background as teaching assistant in different engineering-related fields, enables me to understand and apply analysis rubrics effectively. It also allows me to challenge AI models with advanced technic concepts in statistics, data science, econometrics, fluid dynamics, management, and optimization. During my PhD coursework I had the opportunity to serve as a teaching assistant in deep learning for graduate and profesional education students. This experience provided me with valuable insights about common failures in the data collection and reasoning of IA models, enhancing my expertise in this field. Additionally, it equipped me with the skills in fine-tuning to improve models in image recognition and large language models (LLM).

Entry LevelEnglishSpanish

Labeling Experience

Cypher Evals

Internal Proprietary ToolingTextEvaluation Rating
This project focuses on evaluating model quality by assessing two responses in categories such as writing quality, following instruction, truthfulness, and more. The task requires determining which response is superior and providing an explanation of the criteria considered in the evaluation

This project focuses on evaluating model quality by assessing two responses in categories such as writing quality, following instruction, truthfulness, and more. The task requires determining which response is superior and providing an explanation of the criteria considered in the evaluation

2024

Cypher RHLF

Internal Proprietary ToolingTextText GenerationEvaluation Rating
The primary goal is to evaluate the quality of models by designing prompts -some of which involve generating a trustworthy reference text- intended to challenge the model and expose potential failures. The task includes assessing model responses based on a multidimensional rubric, such as writing quality, adherence to instructions, and truthfulness. After evaluating the models' performance, the process involves selecting the better response to complete the task. In some cases, it is also necessary to correct the errors reported in one of the responses.

The primary goal is to evaluate the quality of models by designing prompts -some of which involve generating a trustworthy reference text- intended to challenge the model and expose potential failures. The task includes assessing model responses based on a multidimensional rubric, such as writing quality, adherence to instructions, and truthfulness. After evaluating the models' performance, the process involves selecting the better response to complete the task. In some cases, it is also necessary to correct the errors reported in one of the responses.

2024

Education

P

Pontificia Universidad Católica

Master of Engineering Science, Strategic Management

Master of Engineering Science
2022 - 2023
P

Pontificia Universidad Católica

Certificate, Mechanics Engineering

Certificate
2018 - 2019

Work History

S

Social System Modelling

Co-Founder and Consultant

Santiago
2024 - Present
I

Independent

Consultant

Santiago
2020 - Present