Jessé Gomes De Almeida

Key Skills

Software

Appen

Clickworker

Label Studio

Lionbridge

Mindrift

OneForma

Scale AI

Toloka

Telus

Other

Top Subject Matter

No subject matter listed

Top Data Types

Text

Audio

Image

Top Task Types

Text Generation

Translation Localization

RLHF

Evaluation Rating

Prompt Response Writing SFT

Classification

Data Collection

Transcription

Segmentation

Object Detection

Red Teaming

Freelancer Overview

I am an AI Training and Data Labeling Specialist with over five years of experience in the digital industry and more than three years working directly with AI training, evaluation, and annotation projects. My work focuses on data annotation, linguistic labeling, and quality assurance for large language models, with an emphasis on accuracy, instruction compliance, and cultural localization for Brazilian Portuguese. I have contributed to projects involving LLM evaluation, RLHF workflows, conversational AI testing, translation and transcription datasets, and multimodal data (text, audio, and image). I regularly collaborate with global teams, applying structured feedback, auditing guidelines, and safety standards to improve model reliability and overall data quality.

Labeling Experience

Image Annotation & Visual Data Evaluation — Freelancer

Scale AIImageSegmentationClassification

Worked as a freelancer on image annotation and visual data evaluation projects for training computer vision models. The work focused on image classification, visual quality assessment, and validation of visual datasets to ensure semantic consistency and suitability for AI training. Responsibilities included reviewing images according to detailed guidelines, identifying visual inconsistencies and edge cases, evaluating content quality and relevance, and assessing potential risks associated with visual data. All activities were performed with an emphasis on annotation accuracy, standardization, and maintaining high quality and safety standards for computer vision training datasets.

Worked as a freelancer on image annotation and visual data evaluation projects for training computer vision models. The work focused on image classification, visual quality assessment, and validation of visual datasets to ensure semantic consistency and suitability for AI training. Responsibilities included reviewing images according to detailed guidelines, identifying visual inconsistencies and edge cases, evaluating content quality and relevance, and assessing potential risks associated with visual data. All activities were performed with an emphasis on annotation accuracy, standardization, and maintaining high quality and safety standards for computer vision training datasets.

2024

Audio Transcription & Linguistic Annotation — Freelancer

Scale AIAudioClassificationTranslation Localization

Worked as a freelancer on Brazilian Portuguese audio transcription projects for training Automatic Speech Recognition (ASR) systems. The work focused on producing accurate and clear transcriptions, ensuring high fidelity to the original audio and proper representation of regional linguistic variations. Responsibilities included transcribing large-scale audio datasets for ASR, applying linguistic review protocols, identifying and standardizing accents across different regions of Brazil, handling informal speech and pronunciation variations, and reviewing and correcting transcriptions according to strict guidelines. All activities were performed with close attention to data integrity, linguistic consistency, and the maintenance of high-quality standards required for use in speech recognition models.

Worked as a freelancer on Brazilian Portuguese audio transcription projects for training Automatic Speech Recognition (ASR) systems. The work focused on producing accurate and clear transcriptions, ensuring high fidelity to the original audio and proper representation of regional linguistic variations. Responsibilities included transcribing large-scale audio datasets for ASR, applying linguistic review protocols, identifying and standardizing accents across different regions of Brazil, handling informal speech and pronunciation variations, and reviewing and correcting transcriptions according to strict guidelines. All activities were performed with close attention to data integrity, linguistic consistency, and the maintenance of high-quality standards required for use in speech recognition models.

2024

AI Training & Data Labeling Specialist — Freelancer

Scale AITextText GenerationTranslation Localization

Worked as both a contributor and a reviewer/auditor in AI training and data labeling projects, handling large-scale textual datasets for the training and evaluation of language models. Responsibilities included data annotation, written localization checks, evaluation of AI-generated responses (including factual accuracy and hallucination detection), instruction-following analysis, linguistic quality assessment, and alignment with safety and quality guidelines. The work involved RLHF workflows, response evaluation and ranking, prompt and response writing and review (SFT), creative writing tasks, as well as adversarial testing and edge case analysis. All activities were conducted in collaboration with global teams, following detailed guidelines, quality metrics, and ethical standards, with a strong focus on continuously improving the performance, reliability, and safety of generative AI systems, particularly for Brazilian Portuguese.

Worked as both a contributor and a reviewer/auditor in AI training and data labeling projects, handling large-scale textual datasets for the training and evaluation of language models. Responsibilities included data annotation, written localization checks, evaluation of AI-generated responses (including factual accuracy and hallucination detection), instruction-following analysis, linguistic quality assessment, and alignment with safety and quality guidelines. The work involved RLHF workflows, response evaluation and ranking, prompt and response writing and review (SFT), creative writing tasks, as well as adversarial testing and edge case analysis. All activities were conducted in collaboration with global teams, following detailed guidelines, quality metrics, and ethical standards, with a strong focus on continuously improving the performance, reliability, and safety of generative AI systems, particularly for Brazilian Portuguese.

2024

Education

Work History