Juliana Carvajal Bautista

Key Skills

Software

Appen

Toloka

Other

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Label Types

Audio Recording

Classification

Entity Ner Classification

Evaluation Rating

Segmentation

Freelancer Overview

Spanish-Fluent Data Annotator with over 2 years of hands-on experience in data labeling, annotation, and AI training projects. Skilled in high-accuracy transcription, segmentation, and classification of Spanish-language audio data, with a proven track record of delivering quality assurance (QA) in annotation projects. Experienced in evaluating search engine results and social media content to ensure compliance with guidelines and maintain data accuracy.

Labeling Experience

Internet Search Evaluator

AppenTextEntity Ner ClassificationClassification

Scope: » Evaluated search engine results, ads, and websites across three project phases, ensuring accurate categorization and compliance with guidelines. » Analyzed text-based content (keywords, metadata, URLs) to determine relevance and intent. Key Tasks: ✔ Taxonomy Classification: Labeled ads, websites, and custom events into specialized categories. ✔ Cross-Lingual Evaluation: Assessed content in multiple languages using translation tools when required. ✔ Intent Analysis: Determined user action/audience purpose from search queries and event data. Project Scale: Processed 800+ tasks weekly (20 hrs/week at ~40 tasks/hour) with consistent adherence to deadlines. Quality Measures: » Recognized as 'Excellent' rater in quality audits, adhering strictly to project guidelines » Completed mandatory calibration tests weekly and guideline updates to ensure rating precision.

Scope: » Evaluated search engine results, ads, and websites across three project phases, ensuring accurate categorization and compliance with guidelines. » Analyzed text-based content (keywords, metadata, URLs) to determine relevance and intent. Key Tasks: ✔ Taxonomy Classification: Labeled ads, websites, and custom events into specialized categories. ✔ Cross-Lingual Evaluation: Assessed content in multiple languages using translation tools when required. ✔ Intent Analysis: Determined user action/audience purpose from search queries and event data. Project Scale: Processed 800+ tasks weekly (20 hrs/week at ~40 tasks/hour) with consistent adherence to deadlines. Quality Measures: » Recognized as 'Excellent' rater in quality audits, adhering strictly to project guidelines » Completed mandatory calibration tests weekly and guideline updates to ensure rating precision.

2023

Social media evaluator

AppenTextEntity Ner ClassificationClassification

Scope: » Evaluated user-generated content across Spanish-language social media platforms to identify policy violations, misinformation, and sensitive topics. » Classified posts/comments based on community guidelines and cultural context (Latin American focus). Key Tasks: ✔ Content Labeling: Tagged posts for hate speech, violence, and fake news using platform-specific criteria. ✔ Data Verification: Cross-checked claims in posts against credible sources when needed. Project Scale: Analyzed 250+ social media posts during 20 hours per week. Quality Measures: Followed Appen’s strict QA protocols and calibration guidelines. Participated in weekly reviews to maintain labeling consistency.

Scope: » Evaluated user-generated content across Spanish-language social media platforms to identify policy violations, misinformation, and sensitive topics. » Classified posts/comments based on community guidelines and cultural context (Latin American focus). Key Tasks: ✔ Content Labeling: Tagged posts for hate speech, violence, and fake news using platform-specific criteria. ✔ Data Verification: Cross-checked claims in posts against credible sources when needed. Project Scale: Analyzed 250+ social media posts during 20 hours per week. Quality Measures: Followed Appen’s strict QA protocols and calibration guidelines. Participated in weekly reviews to maintain labeling consistency.

2024 - 2024

Spanish Audio Transcription Specialist | QA Validator

AppenAudioEntity Ner ClassificationSegmentation

Scope: » Transcribed and labeled Latin American Spanish audio datasets for AI/ML training, ensuring linguistic accuracy and compliance with annotation guidelines. » Served as QA validator to guarantee dataset precision. Key Tasks: ✔ Audio Segmentation: Divided recordings into fixed-duration segments, tagging validity (valid/invalid + subcategories). ✔ Speaker/Context Labeling: Identified gender, speaker turns, background noise, and speech types (e.g., singing, rapping). ✔ Grammatically Correct Transcription: Applied proper orthography and punctuation, correcting mispronunciations while preserving natural speech patterns. ✔ QA Oversight: Audited 20% of team submissions, improving overall accuracy to 98%. Project Scale: Processed 1,500+ audio segments weekly. Quality Measures: » Maintained <2% error rate in transcription/labeling audits. » Reduced invalid segment classifications by 15% through iterative feedback.

Scope: » Transcribed and labeled Latin American Spanish audio datasets for AI/ML training, ensuring linguistic accuracy and compliance with annotation guidelines. » Served as QA validator to guarantee dataset precision. Key Tasks: ✔ Audio Segmentation: Divided recordings into fixed-duration segments, tagging validity (valid/invalid + subcategories). ✔ Speaker/Context Labeling: Identified gender, speaker turns, background noise, and speech types (e.g., singing, rapping). ✔ Grammatically Correct Transcription: Applied proper orthography and punctuation, correcting mispronunciations while preserving natural speech patterns. ✔ QA Oversight: Audited 20% of team submissions, improving overall accuracy to 98%. Project Scale: Processed 1,500+ audio segments weekly. Quality Measures: » Maintained <2% error rate in transcription/labeling audits. » Reduced invalid segment classifications by 15% through iterative feedback.

2023 - 2024

Data Annotation Specialist | AI Training Data

AppenTextEntity Ner ClassificationClassification

Scope: Annotated and labeled diverse datasets (text queries and audio clips) for AI training, focusing on content sensitivity and quality evaluation. Key Tasks: ✔ Sensitive Content Labeling: Categorized user search terms into policy-defined taxonomies (e.g., illegal, explicit, or nonsensical queries) using multi-platform research. ✔ Audio Quality Evaluation: Rated natural and synthetic voices for persuasiveness, emotional tone, and preference metrics. ✔ Guideline Compliance: Ensured consistent labeling. Project Scale: Processed 300+ text weekly tasks.

Scope: Annotated and labeled diverse datasets (text queries and audio clips) for AI training, focusing on content sensitivity and quality evaluation. Key Tasks: ✔ Sensitive Content Labeling: Categorized user search terms into policy-defined taxonomies (e.g., illegal, explicit, or nonsensical queries) using multi-platform research. ✔ Audio Quality Evaluation: Rated natural and synthetic voices for persuasiveness, emotional tone, and preference metrics. ✔ Guideline Compliance: Ensured consistent labeling. Project Scale: Processed 300+ text weekly tasks.

2022 - 2023

Education

Work History