For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

LoginCreate account
Invite to Job
Inaiá Centeio

Inaiá Centeio

LLM Trainer with 4+ years of experience in Portuguese Transcription

PORTUGAL flag
Barcelos, Portugal
$12.00/hrExpertLabelboxLabel StudioOneforma

Key Skills

Software

LabelboxLabelbox
Label StudioLabel Studio
OneFormaOneForma
RemotasksRemotasks
Scale AIScale AI
TelusTelus
Other
Internal/Proprietary Tooling

Top Subject Matter

LLM trainer generalist
Portuguese transcription
Prompt and response writing and rewriting

Top Data Types

AudioAudio
ImageImage
TextText

Top Label Types

Classification
Prompt Response Writing SFT
Question Answering
Text Generation
Text Summarization
Transcription
RLHF
Segmentation

Freelancer Overview

With over 6 years of experience in high-stakes linguistic projects, I specialize in bridging the gap between human language and Artificial Intelligence. My expertise spans LLM Training (RLHF), Prompt Engineering, and Data Annotation, with a deep focus on Brazilian, European, and PALOP Portuguese variants. What I bring to the table: AI & Machine Learning: Specialized in chatbot evaluation, gold-standard labeling, and RLHF, ensuring AI responses are safe, factually accurate, and culturally nuanced. Subject Matter Expertise: Extensive background in Legal (peting, appeals, and judicial proceedings) and Healthcare (Radiological reports and Forensic Medicine), providing high-precision data for specialized industries. Cross-Dialectal Mastery: Expert at navigating the linguistic nuances of PT-BR, PT-EU, and African Portuguese (Mozambique/PALOP). Award-Winning Quality: Recipient of the "Question Master Award" for technical insight and identifying critical project inconsistencies. From transcribing investigative documentaries on international crime to managing large-scale data projects as a Project Manager, my focus is always on high-fidelity delivery and strategic problem-solving. Technical Toolkit: Hotpepper, InqScribe, Sonix, Iara Health, GoTranscript, and Advanced MS Word.

ExpertPortugueseEnglish

Labeling Experience

AI Specialist | Full-cycle AI Training (LLM) | RLHF & Prompt Engineering

OtherTextRLHF
Prompt Engineering & Content Creation: Designed and authored complex prompts across a wide range of domains to stress-test LLM capabilities and improve model responsiveness. Response Editing & Refinement (SFT): Acted as a Lead Editor for AI-generated outputs, correcting hallucinations and refining tone, grammar, and logical flow to create "Golden Responses" for Supervised Fine-Tuning. Multi-Dimensional Classification: Categorized and labeled vast datasets of AI interactions based on intent, safety, and helpfulness, providing the structured data necessary for model alignment. Quality Assurance & Task Review: Served as a Senior Reviewer (L2/L3 level), auditing the work of other annotators to ensure 100% compliance with project guidelines and maintaining high data integrity. Comparative Ranking (RLHF): Evaluated and ranked multiple AI model outputs, providing detailed qualitative rationales to justify the superiority of one response over another.

Prompt Engineering & Content Creation: Designed and authored complex prompts across a wide range of domains to stress-test LLM capabilities and improve model responsiveness. Response Editing & Refinement (SFT): Acted as a Lead Editor for AI-generated outputs, correcting hallucinations and refining tone, grammar, and logical flow to create "Golden Responses" for Supervised Fine-Tuning. Multi-Dimensional Classification: Categorized and labeled vast datasets of AI interactions based on intent, safety, and helpfulness, providing the structured data necessary for model alignment. Quality Assurance & Task Review: Served as a Senior Reviewer (L2/L3 level), auditing the work of other annotators to ensure 100% compliance with project guidelines and maintaining high data integrity. Comparative Ranking (RLHF): Evaluated and ranked multiple AI model outputs, providing detailed qualitative rationales to justify the superiority of one response over another.

2024 - 2025

Linguistic QA & Audio-to-Text Data Annotation

VideoTranscription
Portuguese Audio Data Annotation (PT/BR); Multimedia Localization: Providing subtitling for Chinese productions, ensuring PT-BR output using SubtitleEdit. Subtitle Optimization & QA: Reviewing and adjusting commercial subtitle metadata and timing to improve user experience.

Portuguese Audio Data Annotation (PT/BR); Multimedia Localization: Providing subtitling for Chinese productions, ensuring PT-BR output using SubtitleEdit. Subtitle Optimization & QA: Reviewing and adjusting commercial subtitle metadata and timing to improve user experience.

2022 - 2026

Medical Data Annotation

AudioTranscription
High-Fidelity Medical Transcription: Transcribed clinical dialogues between doctors and patients, covering complex pathologies, pharmacological terms, and specialized medical procedures. Audio Segmentation: Utilized AI software to perform precise audio segmentation and speaker diarization, ensuring perfect alignment between timestamps and medical speech.

High-Fidelity Medical Transcription: Transcribed clinical dialogues between doctors and patients, covering complex pathologies, pharmacological terms, and specialized medical procedures. Audio Segmentation: Utilized AI software to perform precise audio segmentation and speaker diarization, ensuring perfect alignment between timestamps and medical speech.

2025 - 2025

Audio Quality Assurance (QA)/Speech Evaluation

OtherAudioRLHF
AI Audio Quality Evaluation: Conducted rigorous quality assessments of AI-generated speech (Text-to-Speech/TTS), evaluating outputs based on specific project guidelines and technical metrics. Speech Naturalness & Fluency Analysis: Rated audio recordings for prosody, intonation, and human-like flow to identify glitches, robotic artifacts, or synthetic distortions. Guideline Compliance: Applied complex grading scales to measure the helpfulness, safety, and phonetic accuracy of AI voice responses across general topics. Qualitative Feedback & Reasoning: Provided detailed, evidence-based comments for each evaluation, explaining technical discrepancies, tone inconsistencies, or phonetic errors to guide model optimization.

AI Audio Quality Evaluation: Conducted rigorous quality assessments of AI-generated speech (Text-to-Speech/TTS), evaluating outputs based on specific project guidelines and technical metrics. Speech Naturalness & Fluency Analysis: Rated audio recordings for prosody, intonation, and human-like flow to identify glitches, robotic artifacts, or synthetic distortions. Guideline Compliance: Applied complex grading scales to measure the helpfulness, safety, and phonetic accuracy of AI voice responses across general topics. Qualitative Feedback & Reasoning: Provided detailed, evidence-based comments for each evaluation, explaining technical discrepancies, tone inconsistencies, or phonetic errors to guide model optimization.

2025 - 2025

Project Manager & Language Specialist (AI Training)

OtherAudioTranscription
Project Management & Team Leadership: Orchestrated the transcription and review pipeline, managing a team of linguists. Responsible for evaluating transcriber performance and promoting high-performing individuals to Reviewer roles based on quality metrics. Specialized Economic Transcription: Performed high-fidelity transcription and post-editing of complex audio data involving Macroeconomics and Microeconomics, ensuring the correct use of financial and economic nomenclature. Quality Assurance & Compliance: Audited transcripts and reviews to ensure strict adherence to project-specific guidelines, maintaining the highest standards of data integrity for the client.

Project Management & Team Leadership: Orchestrated the transcription and review pipeline, managing a team of linguists. Responsible for evaluating transcriber performance and promoting high-performing individuals to Reviewer roles based on quality metrics. Specialized Economic Transcription: Performed high-fidelity transcription and post-editing of complex audio data involving Macroeconomics and Microeconomics, ensuring the correct use of financial and economic nomenclature. Quality Assurance & Compliance: Audited transcripts and reviews to ensure strict adherence to project-specific guidelines, maintaining the highest standards of data integrity for the client.

2024 - 2024

Education

U

Universidade de Cuiabá - UNIC

Bachelor's Degree, Law

Bachelor's Degree
2010 - 2014

Work History

A

AIxBlock

Transcriber and segmenter

Remote
2025 - 2025
F

First Source

Audio tutor and QA

Remote
2025 - 2025