Tobias Pontelli

Key Skills

Software

Clickworker

Labelbox

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Document

Geospatial Tiled Imagery

Image

Text

Top Label Types

Classification

Question Answering

Mapping

Evaluation Rating

Transcription

Bounding Box

Polygon

Point Key Point

Cuboid

Freelancer Overview

I am a geologist with hands-on experience in data annotation and AI training data quality assurance, gained through roles where I evaluated and validated AI-generated outputs across geospatial and multimodal datasets. My background in geology and field work has given me strong analytical skills and attention to detail, which I apply in reviewing text, spatial data, and complex project guidelines to ensure accuracy, consistency, and reliability. I am skilled in tools such as QGIS, ArcGIS, and various data analysis software, and have contributed to the refinement of evaluation procedures by providing precise feedback and structured tagging. I thrive in collaborative environments, enjoy problem-solving, and am committed to improving the performance and safety of AI systems through meticulous data labeling and annotation.

Labeling Experience

Online Data Analyst

Internal Proprietary ToolingGeospatial Tiled ImageryClassificationQuestion Answering

Evaluation and validation of AI-generated outputs across large-scale data platforms. Evaluation of spatial accuracy, data consistency, and classification of geographic features. Quality assurance (QA/QC) of datasets used for training, testing, and validating machine learning models. Application of technical criteria and project guidelines to ensure data consistency and reliability

Evaluation and validation of AI-generated outputs across large-scale data platforms. Evaluation of spatial accuracy, data consistency, and classification of geographic features. Quality assurance (QA/QC) of datasets used for training, testing, and validating machine learning models. Application of technical criteria and project guidelines to ensure data consistency and reliability

2025

Data Annotator

LabelboxImageBounding BoxPolygon

Performed detailed evaluations and annotations of text and multimodal data. Assessed AI-generated outputs for correctness, coherence, clarity, cultural appropriateness, and safety across diverse tasks and content domains. Identified subtle errors, ambiguities, and biases in model responses, providing precise written feedback and structured tagging to improve model behavior. Consistently applied complex project guidelines, contributing to the refinement of evaluation procedures and escalating ambiguous cases whennnecessary.

Performed detailed evaluations and annotations of text and multimodal data. Assessed AI-generated outputs for correctness, coherence, clarity, cultural appropriateness, and safety across diverse tasks and content domains. Identified subtle errors, ambiguities, and biases in model responses, providing precise written feedback and structured tagging to improve model behavior. Consistently applied complex project guidelines, contributing to the refinement of evaluation procedures and escalating ambiguous cases whennnecessary.

2024 - 2025

Education

Work History