Profile
Roberto is a Mining Engineer graduated from the Federal University of Ouro Preto (Brazil). He holds a MSc and PhD in Geostatistics from the Federal University of Rio Grande do Sul (Brazil) with a focus on implicit geological modeling and uncertainty assess in geological models. Roberto possesses experience in multivariate ore grades estimation and simulation workflows, resources classification and drill hole spacing analysis. He is proficient in Python programming and experienced in applying Machine Learning techniques to geosciences. He currently serves as a Senior geostatistics and data science consultant at Geovariances.
Interactive Geomodeling Sandbox
Explore dynamic spatial estimation based on the Walker Lake repository. Adjust the variogram structural parameters below to watch Ordinary Kriging maps and structural variance update in real-time.
Work Experience
Avon, France - Presential
- Projects management.
- Development of custom geostatistical and machine learning based solutions for the mining industry using Python together with the geostatistical software Isatis.neo.
- Training and mentoring in geostatistics and Python.
- Consulting in geostatistics.
Halmahera, Indonesia - Presential
- Upgrading, maintenance and improvement of the recoverable resource models scripts.
- Assessment and interpretation of data and results.
- Drill hole spacing and resources classification studies.
- Reconciliation program.
- Implementation and proposition of good practices.
- Team training in geostatistics and Python.
Belo Horizonte, Brazil - Remote
- Development of custom geostatistical and machine learning based solutions for the mining industry using Python together with the geostatistical software Isatis.neo.
- Training and mentoring in geostatistics and Python.
- Consulting in geostatistics.
Belo Horizonte, Brazil - Remote
- Development of custom geostatistical and machine learning based solutions for the mining industry using Python together with the geostatistical software Isatis.neo.
Porto Alegre, Brazil - Presential
- Water resources management for mining and energy projects (hydric and thermoelectric).
- Planning and execution of public policies and production of reports and diagnoses with the objective of promoting the mining industry in the state of Rio Grande do Sul.
Denver, USA - Remote
- Production of ar2gas geostatistical software documentation.
- Development of scripts and widgets for the geostatistical software ar2gas.
- Development of custom geostatistical and machine learning based workflows using Python together with the geostatistical software ar2gas for the mining and oil industry.
Porto Alegre, Brazil - Presential
- Research in implicit geologic modeling, uncertainty assessment in geological models, grades estimation and geologic modeling using neural networks and clustering algorithms applied to defining geostatistical domains;
- Teaching assistant of the geostatistics and machine learning classes of the Graduate Program in Mining, Metallurgical and Materials Engineering at UFRGS;
- Professor of the Python Applied to Geostatistics class of the Graduate Program in Mining, Metallurgical and Materials Engineering at UFRGS;
- Taught Python and geostatistics courses in company.
Featured Projects
An Android app to log drill holes and upload data to a server, streamlining field data collection for geological teams. [View on GitHub]
Implementation in Python for the simplified methodology to define the flood spot in the event of a dam failure. [View on GitHub]
Education
Conclusion: july 2021
Thesis: Geologic modeling and uncertainty assess using signed distances functions
Conclusion: january 2017
Dissertation: Implicit geologic modeling using signed distances functions
Conclusion: september 2013
Undergraduate thesis: Study of technical aspects essential to the evaluation of mineral reserves
Interests
Talks & Presentations
Machine learning webinar - Geovariances (English)
Defining Geometalurgical Domains in a Phosphate Mine Using Machine Learning - GEOMET 2022 (English)