Matías C.

Key Skills

Software

Labelbox

Scale AI

Google Cloud Vertex AI

Top Subject Matter

Product Data Classification

Top Data Types

Text

Document

Top Task Types

Text Generation

Fine-tuning

Classification

Freelancer Overview

Senior AI Engineer with 10 years of experience designing AI, ML, and data systems for enterprisescale use cases across Scale AI, Surge AI, and DataAnnotation. Builds LLM evaluation pipelines, RAG workflows, prompt libraries, and production data pipelines in Python, SQL, PyTorch, TensorFlow, Hugging Face, AWS, and Kubernetes. Delivered benchmark datasets, synthetic training corpora, and quality scoring systems that improved model accuracy, reduced hallucinations, and accelerated delivery for cross-functional teams. Strong record in MLOps, cloud architecture, distributed systems, and deploying testable AI services with measurable gains in latency, throughput, and evaluation coverage.

Labeling Experience

Senior AI Engineer

LabelboxTextFine-tuning

As a Senior AI Engineer at Scale AI, I designed and managed LLM training and annotation workflows for improving NLP model performance. I refined prompt engineering, established annotation guidelines, and standardized labeling practices across multi-platform tools. I analyzed model outputs for errors, performed data quality audits, mentored trainers on review methods, and optimized model evaluation routines. • Led the creation of training datasets and edge-case annotation standards using Labelbox and Scale AI. • Conducted A/B and benchmarking tests to validate prompt and labeling improvements. • Developed QA and data audit scripts for large-scale training data reviews in Python and SQL. • Enhanced model fairness, safety, and consistency through targeted dataset design and bias monitoring.

2022 - Present

AI Engineer

TextFine-tuning

At Surge AI, I designed, processed, and reviewed annotated text datasets to train and validate AI models. I evaluated NLP outputs, established and maintained custom labeling benchmarks, and executed QA for label drift and error modes. I contributed to prompt template creation and audit routines to improve classification accuracy and consistency. • Used spreadsheets and Python to maintain high annotation quality and reviewer notes. • Defined labeling acceptance criteria and operationalized reporting for continuous model improvement. • Performed structured error analysis for edge cases, bias, and hallucination. • Built and tested NLP training workflows leveraging TensorFlow and SQL.

2018 - 2022

AI Trainer

TextClassification

Supported the labeling and review of product data within Python-assisted workflows to enhance AI classification systems. Validated generated AI outputs against specific written guidelines and requirements to ensure consistency and accuracy. Used Pandas to identify anomalies in datasets before training cycles commenced. • Labeled over 8,000 product data records to support AI content quality improvement. • Evaluated outputs for assumptions, completeness, and compliance with project standards. • Provided feedback on ambiguous and edge cases to reduce rework and improve annotation consistency. • Collaborated across teams to refine review criteria and improve data labeling standards.

2016 - 2018

Education

Work History