Geophysical Engineering student & computational developer specializing in applied machine learning, subsurface signal processing, and scientific software engineering.
Desktop Petrophysical Analysis Suite
A full-featured PyQt5 desktop application for oil & gas well-log evaluation. Features 9 calculation modules (Vsh, Porosity, Sw, Permeability, Net Pay, Pickett Plot), multi-format LAS/Excel export, and an interactive Matplotlib QC tool for curve despiking, smoothing, and interpolation.
Privacy-First In-Browser Media & Dev Utilities
A high-performance web platform bundling 30+ client-side tools for AI background removal, image optimization, PDF processing (merge/split/OCR), media transcoding via FFmpeg WASM, and developer utilities. All computation runs 100% in-browser without server uploads.
ATS-Optimized Resume Engine & Heuristic Analyzer
A web platform for crafting parseable resumes with real-time A4 pagination (never splitting blocks across page breaks). Includes a 0–100 ATS scoring engine, job description keyword matcher, 3 single-column templates, and local-first JSON project import/export.
AI Voice Tutor & Language Learning Android App
An offline-first native Android application with Jetpack Compose, Room, and Hilt. Integrates Qwen LLM for real-time speech conversation, speech recognition (ASR), pronunciation scoring, spaced-repetition flashcards, and background daily study reminders across 6 languages.
Neural Signal Enhancement Pipeline
End-to-end PyTorch deep learning framework utilizing CNN and LSTM architectures to filter heavy electromagnetic cultural noise from Magnetotelluric field data, boosting signal-to-noise ratio by 58.8% with a 3x runtime speedup over conventional filtering baselines.
Missing Curve Synthesis Engine
Machine learning modeling pipeline combining LightGBM and Bidirectional RNNs to synthesize missing subsurface petrophysical logs across oil fields, outperforming baseline models by up to 6.5x in predictive accuracy.
Automated Reservoir Processing Toolkit
An interactive Streamlit application built on top of tNavigator seismic interpretation workflows (well-seismic tie, horizon picking, seismic attributes) to streamline data visualization and automate new-well petrophysical processing for the Kupe South field.
Journal of Human, Earth, and Future, 2025
READ JOURNAL ARTICLE (HEF) ↗2025 International Conference on Smart Computing, IoT and Machine Learning (SIML), IEEE, 2025
READ PROCEEDINGS (IEEE XPLORE) ↗My work focuses on bridging physics-based geoscience with modern deep learning and robust software engineering.
From training neural architectures on noisy subsurface measurements to engineering modular desktop suites for petrophysicists and mentoring future engineers, I prioritize numerical precision, modular design, and reliable performance.
LET'S CONNECT & COLLABORATE.