Cecilia Sanchez

Key Skills

Software

Clickworker

CVAT

OneForma

Remotasks

Scale AI

Toloka

Internal/Proprietary Tooling

Other

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Video

Top Task Types

Bounding Box

Cuboid

Evaluation Rating

Mapping

Polygon

Freelancer Overview

I have worked in translation and proofreading, and I also have extensive experience writing and transcribing various texts. This experience has allowed me to evaluate and generate prompts and responses for specific LLM projects. I have performed various tasks in graphic design and data entry for websites. I have contributed to projects involving image segmentation, image mapping, image and video annotation for self-driving cars, as well as product classification and text annotation.

Labeling Experience

Audio-to-Text Transcription Specialist (Spanish & Portuguese)

OtherAudioTranslation Localization

I worked as an audio-to-text transcriber for Spanish and Portuguese projects within an artificial intelligence platform. I used the platform’s proprietary software to produce accurate transcriptions, ensuring text quality through grammatical review and meeting established deadlines. I contributed to multilingual projects while maintaining high standards of accuracy and linguistic consistency.

I worked as an audio-to-text transcriber for Spanish and Portuguese projects within an artificial intelligence platform. I used the platform’s proprietary software to produce accurate transcriptions, ensuring text quality through grammatical review and meeting established deadlines. I contributed to multilingual projects while maintaining high standards of accuracy and linguistic consistency.

2023

Visual Annotation of Road Damage and Surface Conditions

Scale AIVideoClassification

This visual annotation project focused on classifying road surface conditions through detailed, frame-by-frame analysis of video frames. My primary responsibility was to accurately identify and delineate damaged sections of the road, marking from where each deterioration began to where it ended. In cases where multiple types of damage—such as cracks, potholes, or other imperfections—were present, I created separate annotations for each one until the entire frame was fully labeled. Additionally, I classified the surface condition of the asphalt, indicating whether it was clean, dirty, or covered with snow. This information was essential for autonomous driving systems to interpret the road environment more accurately and adapt their behavior to varying conditions, enabling safer and more efficient navigation.

This visual annotation project focused on classifying road surface conditions through detailed, frame-by-frame analysis of video frames. My primary responsibility was to accurately identify and delineate damaged sections of the road, marking from where each deterioration began to where it ended. In cases where multiple types of damage—such as cracks, potholes, or other imperfections—were present, I created separate annotations for each one until the entire frame was fully labeled. Additionally, I classified the surface condition of the asphalt, indicating whether it was clean, dirty, or covered with snow. This information was essential for autonomous driving systems to interpret the road environment more accurately and adapt their behavior to varying conditions, enabling safer and more efficient navigation.

2024 - 2024

Vehicle Damage Annotation Specialist

Scale AIImageSegmentation

I participated in an annotation project focused on vehicle damage detection. My task involved visually identifying and segmenting dents and missing parts on the exterior of vehicles using precise polygon-based segmentation on images. This process helped generate high-quality data to train computer vision models used for automated vehicle inspection.

I participated in an annotation project focused on vehicle damage detection. My task involved visually identifying and segmenting dents and missing parts on the exterior of vehicles using precise polygon-based segmentation on images. This process helped generate high-quality data to train computer vision models used for automated vehicle inspection.

2024 - 2024

AI Prompt Engineer & Response Evaluator

Scale AITextPrompt Response Writing SFT

In this project, I created prompts with specific guidelines based on assigned categories and subcategories, as well as varying difficulty levels to guide content generation. My task involved evaluating two responses generated by artificial intelligence, selecting the better one, and if the chosen response lacked relevant information for the prompt, rewriting it to improve its quality. Additionally, depending on the number of turns required by the task, I continued the conversation while ensuring coherence and adherence to instructions. Finally, I rated the evaluated response on key aspects such as clarity, grammar, factual accuracy, verbosity level, and compliance with general instructions. This work demanded analytical skills, strong language proficiency, and a deep understanding of human-AI interaction.

In this project, I created prompts with specific guidelines based on assigned categories and subcategories, as well as varying difficulty levels to guide content generation. My task involved evaluating two responses generated by artificial intelligence, selecting the better one, and if the chosen response lacked relevant information for the prompt, rewriting it to improve its quality. Additionally, depending on the number of turns required by the task, I continued the conversation while ensuring coherence and adherence to instructions. Finally, I rated the evaluated response on key aspects such as clarity, grammar, factual accuracy, verbosity level, and compliance with general instructions. This work demanded analytical skills, strong language proficiency, and a deep understanding of human-AI interaction.

2024 - 2024

AI Response Quality Analyst

Scale AITextEvaluation Rating

I participated in a content evaluation project where my task was to analyze user-written prompts along with two responses generated by artificial intelligence. My main role was to compare both responses and select the better one based on specific quality criteria. Once the best response was chosen, I had to evaluate it in detail according to several key parameters: overall quality, conversational tone, local adaptation (cultural or contextual relevance), level of verbosity or length, precise adherence to the instructions given in the prompt, as well as the truthfulness and safety of the content. This work required critical thinking, linguistic skills, and an understanding of language model behavior to ensure the AI provided useful, clear, and user-aligned answers.

I participated in a content evaluation project where my task was to analyze user-written prompts along with two responses generated by artificial intelligence. My main role was to compare both responses and select the better one based on specific quality criteria. Once the best response was chosen, I had to evaluate it in detail according to several key parameters: overall quality, conversational tone, local adaptation (cultural or contextual relevance), level of verbosity or length, precise adherence to the instructions given in the prompt, as well as the truthfulness and safety of the content. This work required critical thinking, linguistic skills, and an understanding of language model behavior to ensure the AI provided useful, clear, and user-aligned answers.

2024 - 2024

Show more

Education

Work History