Ricardo M.

Key Skills

Software

No software listed

Top Subject Matter

Image classification

deep learning

LLM Retrieval-Augmented Generation

Top Data Types

Image

Text

Top Task Types

Classification

Freelancer Overview

Image Classification Model Optimization Data Labeling and AI Training. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Engineering, Shenzhen University (2025). AI-training focus includes data types such as Image and Text and labeling workflows including Classification.

Labeling Experience

Text Data Annotation and LLM Training for AI Knowledge Assistant

TextClassification

Curated, chunked, and annotated domain-specific text data for AI knowledge retrieval tasks supporting an LLM-based assistant. Extracted metadata and semantic features from technical documents, textbooks, and past exam papers to train and evaluate retrieval-augmented models. Assessed and classified text segments for relevance, knowledge coverage, and context coherence in support of question-answering and knowledge assistance features. • Labeled and extracted meaningful sections from computer science educational texts. • Implemented semantic similarity retrieval and metadata annotation strategies. • Supported LLM fine-tuning and evaluation with high-quality text data annotation. • Used PGvector and proprietary databases for text data labeling and management.

2024 - Present

Image Classification Model Optimization Data Labeling and AI Training

ImageClassification

Optimized image classification models using the CIFAR-10 dataset by preparing data, selecting model architectures, and applying advanced training strategies. Labeled and cleaned image data, performed data augmentation, and implemented label smoothing to enhance model accuracy. Developed and fine-tuned classification models, integrating multiple regularization and optimization techniques to boost few-shot learning performance. • Built and preprocessed image datasets for model training. • Applied data augmentation, label smoothing, and whitening preprocessing techniques. • Validated and improved image model classification accuracy through systematic annotation and optimization. • Utilized PyTorch and internal tooling to complete end-to-end data labeling for image tasks.

2023 - 2024

Education

Work History