Miguel Garcia

Key Skills

Software

Data Annotation Tech

Mindrift

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Text Summarization

Evaluation Rating

Prompt Response Writing SFT

Freelancer Overview

I am an experienced AI Trainer and Data Annotator with a strong background in supporting large-scale AI systems, particularly in interactive, narrative-driven, and gaming environments. My expertise includes annotating and reviewing complex datasets involving dialogue, character behavior, quests, and world-building elements, as well as evaluating AI-generated responses for quality, tone, and narrative consistency. I have successfully collaborated with cross-functional teams to improve model accuracy and user engagement, and have contributed to both gaming and non-gaming domains, such as customer experience and behavioral analysis. My approach combines analytical rigor with creative insight, and I am recognized for my attention to detail, guideline adherence, and ability to deliver constructive feedback that optimizes AI performance. With a deep passion for story-focused technology and human-centered AI, I am committed to delivering high-quality annotation and evaluation services that drive better model outcomes.

Labeling Experience

data labeller

MindriftImageText SummarizationEvaluation Rating

This project involved large-scale data labeling to support machine learning model training and evaluation. The scope included annotating and validating datasets across multiple data types, with tasks such as classification, tagging, bounding boxes, segmentation, and quality review, depending on project requirements. The project handled a high volume of data items under strict guidelines, ensuring consistency and accuracy across annotations. Quality measures included multi-stage reviews, adherence to detailed labeling instructions, inter-annotator agreement checks, and regular feedback loops to maintain high precision and reliability throughout the dataset.

This project involved large-scale data labeling to support machine learning model training and evaluation. The scope included annotating and validating datasets across multiple data types, with tasks such as classification, tagging, bounding boxes, segmentation, and quality review, depending on project requirements. The project handled a high volume of data items under strict guidelines, ensuring consistency and accuracy across annotations. Quality measures included multi-stage reviews, adherence to detailed labeling instructions, inter-annotator agreement checks, and regular feedback loops to maintain high precision and reliability throughout the dataset.

2021 - 2025

Education

Work History