Sobre

We are developing a prototype system for subsurface resource detection (water and minerals) using passive gravity/seismic-like sensor data. The goal is to analyze time-series signals to identify anomalies, estimate densities, and detect specific material signatures (e.g., water peaks and mineral densities). We need an experienced freelancer to build and refine the data processing pipeline from the ground up, including signal analysis, feature extraction, visualization, and progression toward machine learning-based detection. Key Tasks Batch process large folders of CSV time-series sensor data (using pandas, numpy, scipy). Implement FFT (Fast Fourier Transform) for spectral analysis, noise filtering, peak detection, and amplitude/frequency extraction. Develop models to estimate subsurface densities and anomalies correlated with water or mineral presence. Create clear visualizations (Matplotlib/Plotly) and summary reports or anomaly maps. Optionally: Build machine learning models (scikit-learn or PyTorch) for classifying anomalies across datasets. Optimize code for performance, scalability, and add documentation/error handling. Start with a small paid test task (process sample data and demonstrate FFT-based peak detection for material signatures). Required Skills & Experience Expert in Python (pandas, numpy, scipy, matplotlib). Strong background in signal processing (FFT, spectral analysis, peak detection). Experience with time-series data analysis and anomaly detection. Solid machine learning knowledge (scikit-learn; bonus for TensorFlow/PyTorch). Bonus: Experience in geophysics, remote sensing, seismic/gravity data, exploration, or sensor-based resource detection. Portfolio/GitHub with relevant projects (e.g., sensor data processing, geophysical analysis, Kaggle competitions). Excellent communication and ability to explain technical decisions.