About
I am Dr. Muhammad Amar Gul, a geoscientist and applied AI/ML researcher working at the intersection of economic geology, mineral geochemistry, GeoAI, spatial data science, remote sensing, and mineral exploration.
My work is centered on a practical scientific question: how can complex geoscience data be converted into useful exploration evidence without losing geological meaning, uncertainty, or validation discipline? I develop research and decision-support workflows that connect mineral-system knowledge with machine learning, spatial analysis, geostatistics, remote sensing, and reproducible scientific software.
Professional identity
I work across research, exploration delivery, and scientific software development rather than treating them as separate disciplines. The common thread is translating heterogeneous geological evidence into defensible inference and operational decisions.
My current work includes:
- mineral-system and geochemical interpretation — sulfide mineral chemistry, trace elements, S-Pb isotopes, critical metals, metallogenic discrimination, and ore-system reasoning;
- GeoAI and scientific machine learning — anomaly detection, classification, grouped validation, explainability, uncertainty analysis, and reproducible evaluation;
- spatial exploration intelligence — prospectivity, remote sensing, geophysics, geological and structural evidence integration, geostatistics, GIS, and target prioritization;
- research engineering — Python-based pipelines, ETL, QA gates, automation, batch execution, scientific applications, and auditable technical delivery.
Current appointments
Senior Geoscientist (Data & AI) — Erity Pty Ltd.
Remote, Australia · 2026–present
Research-to-product GeoAI for mineral exploration, including geochemical anomaly intelligence, target detection, multimodal exploration workflows, uncertainty products, geostatistical outputs, map QA, and field-ready decision support.
Project Geologist (AI & ML) — China National Geological & Mining Corporation
Saudi Arabia · 2024–present
Arabian Shield prospectivity and mineral-potential work integrating geology, geochemistry, geophysics, structures, remote sensing, GIS, exploration screening, and geological data systems.
Founder & Director — AIMEX Lab
AIMEX is an independent research initiative focused on artificial intelligence for mineral exploration, reproducible benchmarking, collaborative GeoAI research, and scientifically defensible exploration methodology.
Research trajectory
Mineral geochemistry and ore systems
My geological foundation developed through field and research work on ore deposits, sedimentary systems, mineral chemistry, isotopes, critical metals, and metallogenic interpretation.
Global geochemical AI research
My PhD research at the University of Science and Technology of China combined global pyrite, galena, and sphalerite geochemistry with machine learning to investigate deposit discrimination, tectonic setting, ore genesis, and critical-metal enrichment.
Operational GeoAI and exploration decision systems
I now extend that research discipline into regional prospectivity, anomaly intelligence, multimodal evidence integration, geostatistics, software systems, and exploration decision support.
Selected evidence of work
My research has also contributed to work describing the first reported Ge-rich deposit in Pakistan, linking mineral geochemistry and critical-metal interpretation with broader exploration significance.
Scientific approach
I treat prediction, geological interpretation, and decision support as related but distinct tasks. My preferred workflow is to:
- establish data authority, provenance, exclusions, and QA/QC before modelling;
- preserve geological and spatial context during feature construction;
- use grouped, spatially appropriate, or otherwise defensible validation rather than convenient random splits;
- quantify uncertainty and test sensitivity and transferability where feasible;
- use explainability to support interpretation, not as proof of causality;
- report negative, unstable, or non-transferable results rather than force a positive conclusion.