Abraham Richart Rueda

Key Skills

Software

Labelbox

Mindrift

Scale AI

Toloka

Other

Internal/Proprietary Tooling

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Text

Top Task Types

Evaluation Rating

Prompt Response Writing SFT

RLHF

Text Generation

Translation Localization

Freelancer Overview

I am an experienced AI Trainer and Linguist specializing in the evaluation and development of Large Language Models (LLMs). Since 2023, I’ve worked on multiple high-impact projects involving Reinforcement Learning from Human Feedback (RLHF), prompt engineering, factuality assessments, and red teaming. With a strong academic background in philology and theology and multilingual fluency in English, German, and Spanish, I bring deep linguistic insight, cultural awareness, and precision to AI training workflows. My strengths lie in evaluating AI-generated content for factual accuracy, coherence, and language quality. I’ve contributed to training LLMs by identifying subtle language issues, classifying factual errors by severity, and providing structured feedback to improve model output. I am proficient with tools like Labelbox, Grammarly, and proprietary labeling platforms, and I consistently deliver high-quality, scalable annotation work across domains.

Labeling Experience

Prompt-Response Comparison & Ranking (RLHF)

MindriftTextRLHFEvaluation Rating

Ranked multiple model-generated responses to prompts based on accuracy, helpfulness, coherence, and stylistic quality. Applied detailed rubric-based assessments to support reinforcement learning through human feedback (RLHF). Helped fine-tune model behavior by providing comparative judgments and constructive analysis to guide reward modeling.

Ranked multiple model-generated responses to prompts based on accuracy, helpfulness, coherence, and stylistic quality. Applied detailed rubric-based assessments to support reinforcement learning through human feedback (RLHF). Helped fine-tune model behavior by providing comparative judgments and constructive analysis to guide reward modeling.

2023

LLM Evaluation – Factuality, Red Teaming & Prompt Optimization

Scale AITextQuestion AnsweringRLHF

In this project, I evaluated outputs from large language models (LLMs) with a focus on factual accuracy, harmful content, and linguistic quality. Tasks included verifying facts via web research, classifying errors (e.g., inaccurate, unsupported, or disputed), and rating responses by relevance, completeness, and tone. I also contributed to safety audits through red teaming—generating adversarial prompts to test model robustness. Additionally, I performed comparative ranking of AI outputs and participated in fine-tuning datasets by writing high-quality prompt-response pairs in English, German, and Spanish. My contributions supported RLHF pipelines and safety improvements for production-ready AI systems. I maintained strict adherence to evaluation rubrics and quality standards and consistently ranked in the top tier of annotators for accuracy and attention to detail.

In this project, I evaluated outputs from large language models (LLMs) with a focus on factual accuracy, harmful content, and linguistic quality. Tasks included verifying facts via web research, classifying errors (e.g., inaccurate, unsupported, or disputed), and rating responses by relevance, completeness, and tone. I also contributed to safety audits through red teaming—generating adversarial prompts to test model robustness. Additionally, I performed comparative ranking of AI outputs and participated in fine-tuning datasets by writing high-quality prompt-response pairs in English, German, and Spanish. My contributions supported RLHF pipelines and safety improvements for production-ready AI systems. I maintained strict adherence to evaluation rubrics and quality standards and consistently ranked in the top tier of annotators for accuracy and attention to detail.

2023

German and English Fluency Review

TelusTextText GenerationTranslation Localization

Reviewed AI-generated content in both German and English with a focus on grammar, fluency, and stylistic appropriateness. Delivered corrections and constructive feedback on sentence structure, tone, and register. Specialized in refining content across academic, technical, and conversational styles to align with native-level usage.

Reviewed AI-generated content in both German and English with a focus on grammar, fluency, and stylistic appropriateness. Delivered corrections and constructive feedback on sentence structure, tone, and register. Specialized in refining content across academic, technical, and conversational styles to align with native-level usage.

2022

Education

Work History