Automated NDVI Monitoring & Alert Framework
Computer Vision Engineer
Engineered an object-oriented image processing pipeline in Python using OpenCV to detect early signs of plant water stress through NDVI (Normalized Difference Vegetation Index) trends.
Applied advanced computer vision techniques including lighting normalization, channel extraction (NIR and red), and irrigation-zone-based segmentation to enable pixel-level NDVI trend monitoring.
Conducted hands-on plant experiments to validate the pipeline's accuracy, iteratively testing and refining the system based on observed physiological responses, and integrated a custom API to trigger real-time alerts for deployment in BioPod Model One and upcoming Model Orion.
BioPod Simulation & ML Integration
Autonomous Systems Simulation Engineer
Engineered a physics-based simulation in Python using object-oriented design to model system behavior for the next-generation BioPod, integrating custom machine learning models to power autonomous decision-making.
Collaborated with research engineers to map the energy profiles of individual components, enabling real-time predictions of energy consumption and environmental response under variable conditions.
Analyzed historical weather and internal BioPod data to uncover temperature-driven patterns, optimizing energy efficiency and creating an interactive interface to test adaptive behavior under simulated sensor inputs.
Computer Vision for Plant Growth Analysis
Computer Vision & Plant Phenomics Engineer
Developed monocular depth estimation algorithms via PyTorch's MiDaS to measure vertical plant growth from 2D imagery of a single aerial camera.
Created mathematical models using pixel values of stable reference points within each frame to establish consistent spatial calibration across images.
Calculated and analyzed the evolution of reference points and plant height over time, enabling precise virtual growth monitoring and feedback for BioPod systems.
End-to-End NDVI Data API & Trend Analysis Engine
Agricultural Data Systems Engineer
Designed and deployed an analytical API to monitor NDVI trends from live camera feeds and trigger anomaly alerts.
Built a robust RESTful API to store, retrieve, and manage processed vegetation data in an optimized SQLite backend.
Collaborated closely with plant and engineering teams to ensure proactive detection of system issues, preventing unnoticed failures and plant health decline.
Climate-Responsive Modeling for BioPod Optimization
Climate Data Scientist
Developed multivariate time series models to assess the impact of exogenous variables on BioPod Model One performance.
Built connections between internal BioPod temperature and external weather patterns using SARIMAX and Hidden Markov models, forecasting how varying global climates would impact system performance across multiple cities
Identified critical climate-growth relationships, driving targeted product design improvements based on actionable, data-driven insights.
YOLO-Based Plant Growth & Flowering Detection System
Plant Phenomics & ML Engineer
Trained and fine-tuned custom YOLO object detection models to classify plant growth stages and detect flowering onset for optimized harvest timing.
Manually annotated training datasets and applied image augmentation techniques to enhance model robustness and accuracy.
Developed an automated alert system triggered by flower detection to support research on optimal harvesting schedules and ensure efficient monitoring of plant developmental progress.
OSINT Tools for Crime Investigation & Law Enforcement Support
OSINT Developer & Digital Forensics Engineer
Leveraged GitHub, Python, and command-line scripting to develop sophisticated OSINT tools, enabling efficient extraction, processing, and analysis of social media data to support criminal investigations and provide actionable intelligence.
Authored an instructional handbook for Detectives and Crime Analysts, detailing the technical use of OSINT tools, significantly enhancing team capabilities and streamlining investigative processes.
Designed and developed a user-friendly website to simplify the access and use of OSINT resources, enabling efficient data retrieval and accelerating investigative workflows.
Crime Prediction Machine Learning Project
Data Scientist
Developed multiple machine learning models to predict the likelihood of an individual being arrested for homicide, based on time of year and location, using the Chicago Crime Statistics dataset.
Applied various machine learning algorithms, including Linear Discriminant Analysis, Logistic Regression, K-Nearest Neighbors, and Random Forest, to analyze crime patterns.
Leveraged R for data visualization and model testing, enhancing understanding of key factors influencing crime rates.
Sales Forecasting and Demographic Profiling for E-Commerce
E-Commerce Data Analyst
Developed Python scripts to track and aggregate sales data by zip code, facilitating geographic performance analysis and segmentation.
Applied statistics to analyze sales data and generate target demographic profiles for future marketing initiatives.
Utilized Python libraries such as Pandas, NumPy, and Matplotlib to preprocess data, visualize trends, and extract actionable insights to drive marketing decisions.
Forecasting Solar-Driven Radio Frequency Signals
Time Series Forecasting Analyst
Engineered predictive models for monthly Critical Radio Frequencies influenced by solar and atmospheric variation, optimizing signal reliability for long-range communication.
Performed seasonal decomposition and multi-lag differencing to extract underlying trends, cyclical patterns, and reduce noise in high-variance time series data.
Tuned SARIMA models via AICc, stationarity testing, and residual diagnostics, achieving highly accurate long-term forecasts with robust cross-validation
Quantifying Global Well-Being Trends
Social Data & Insights Analyst
Analyzed global happiness data (2005-2022) to quantify the impact of health, generosity, freedom, and corruption on economic performance across nations.
Used Python to isolate and visualize variable influence on log GDP per capita, building intuitive correlation matrices and trend graphs.
Identified key cross-national drivers of well-being, positioning happiness and mental health as viable metrics for policy design and reform.