Andréa Amancio

Key Skills

Software

Appen

Labelbox

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Image

Text

Video

Top Label Types

Computer Programming Coding

Data Collection

Evaluation Rating

Prompt Response Writing SFT

RLHF

Segmentation

Text Generation

Transcription

Translation Localization

Freelancer Overview

I am a bilingual French computer science student in my final year, with over 15 months of hands-on experience as an AI trainer and data annotator. My work includes evaluating and comparing large language model outputs, performing coding reviews and quality assurance, and conducting AI safety assessments to ensure compliance with ethical guidelines. I have contributed to speech-to-text transcription and translation (EN–FR), as well as multimodal AI training by creating high-quality golden responses for image–prompt pairs. My strong STEM background, fluency in English, and ability to manage remote, part-time roles enable me to deliver precise, reliable training data across NLP and computer vision projects. I am passionate about artificial intelligence and committed to supporting the development of safe, high-quality AI systems.

Labeling Experience

LLM Coding Review / QA

Scale AIComputer Code ProgrammingEvaluation Rating

I reviewed and validated peer evaluations and model comparisons in coding contexts. My role ensured quality and consistency of review outcomes. I provided feedback to improve evaluation standards. • Checked code comparison accuracy • Identified discrepancies • Maintained project quality standards • Collaborated on coding review protocols

I reviewed and validated peer evaluations and model comparisons in coding contexts. My role ensured quality and consistency of review outcomes. I provided feedback to improve evaluation standards. • Checked code comparison accuracy • Identified discrepancies • Maintained project quality standards • Collaborated on coding review protocols

2025 - 2024

Multimodal AI Training

Scale AIImagePrompt Response Writing SFT

I wrote high-quality golden responses for image-prompt pairs in a multimodal AI training context. My responses served as reference standards for model training. I ensured clarity and alignment with project guidelines. • Generated reference answers to image prompts • Followed multimodal instruction protocols • Reviewed and corrected draft responses • Contributed to improved image–text dataset quality

I wrote high-quality golden responses for image-prompt pairs in a multimodal AI training context. My responses served as reference standards for model training. I ensured clarity and alignment with project guidelines. • Generated reference answers to image prompts • Followed multimodal instruction protocols • Reviewed and corrected draft responses • Contributed to improved image–text dataset quality

2025 - 2025

Speech-to-Text Transcription & Translation (EN–FR)

AppenAudioTranslation Localization

I transcribed and translated speech data from English to French. My work included quality checking outputs by peers. I ensured the accuracy and integrity of multilingual audio transcriptions. • Processed speech-to-text data • Performed bilingual translation tasks • Conducted peer review of transcriptions • Maintained quality control over outputs

I transcribed and translated speech data from English to French. My work included quality checking outputs by peers. I ensured the accuracy and integrity of multilingual audio transcriptions. • Processed speech-to-text data • Performed bilingual translation tasks • Conducted peer review of transcriptions • Maintained quality control over outputs

2025 - 2025

AI Safety Evaluation

Scale AITextEvaluation Rating

I reviewed model responses for harmful content and validated appropriate refusal (punt) behavior. My evaluations ensured adherence to AI safety guidelines. I assessed rule compliance and flagged safety issues. • Implemented safety and refusal protocols • Assessed harmfulness of generated content • Documented findings in evaluation reports • Recommended improvements for model behavior

I reviewed model responses for harmful content and validated appropriate refusal (punt) behavior. My evaluations ensured adherence to AI safety guidelines. I assessed rule compliance and flagged safety issues. • Implemented safety and refusal protocols • Assessed harmfulness of generated content • Documented findings in evaluation reports • Recommended improvements for model behavior

2025 - 2025

LLM Model Comparison (Coding)

Scale AIComputer Code ProgrammingEvaluation Rating

I evaluated and compared outputs from two large language models focused on coding tasks. My work prioritized correctness, logical consistency, and code quality. I systematically identified differences in performance and documented insights. • Conducted head-to-head model comparisons • Analyzed and rated code outputs • Documented methodology • Provided structured feedback to research teams

I evaluated and compared outputs from two large language models focused on coding tasks. My work prioritized correctness, logical consistency, and code quality. I systematically identified differences in performance and documented insights. • Conducted head-to-head model comparisons • Analyzed and rated code outputs • Documented methodology • Provided structured feedback to research teams

2025 - 2025

Education

Work History