I'm Kosol, a Data Science student at the Institute of Technology of Cambodia with a mission: bringing AI from satellites to rice fields.
Growing up in Cambodia, I've seen firsthand how technology can transform livesβespecially in agriculture. That's why I'm dedicated to building practical AI solutions that help farmers, protect crops, and contribute to food security in Southeast Asia.
class DataScientist:
def __init__(self):
self.name = "Kosol Chou"
self.role = "Data Science Student & ML Engineer"
self.location = "Phnom Penh, Cambodia π°π"
self.education = "B.Sc. Data Science @ ITC (2022-Present)"
def current_mission(self):
return "Building AI solutions for agricultural challenges"
def daily_stack(self):
return {
"languages": ["Python", "SQL", "R"],
"ml_frameworks": ["TensorFlow", "scikit-learn", "XGBoost"],
"data_tools": ["Pandas", "NumPy", "Matplotlib"],
"special_skills": ["Remote Sensing", "GIS", "Time Series"]
}πΎ CropXcel - My flagship project combining satellite remote sensing + ML to help farmers prevent crop waterlogging
π Stock Prediction Models - Deep learning with LSTM/GRU for financial forecasting
π Traffic Forecasting - ARIMA-based urban planning solutions
π Continuous Learning - Currently mastering deep learning and advanced time series analysis
Bridging the gap between space technology and sustainable farming
The Challenge: Waterlogging destroys millions of hectares of crops globally, affecting food security and farmer livelihoods.
My Solution: A full-stack web platform that processes Sentinel-1 SAR satellite data in real-time to detect waterlogging risks and provide actionable insights.
| Feature | Technology | Impact |
|---|---|---|
| π°οΈ Real-time Satellite Analysis | Sentinel-1 SAR, Python | Automated detection of waterlogging hotspots |
| πΊοΈ Interactive GIS Dashboard | Leaflet.js, PostgreSQL | Field boundary drawing & multi-layer visualization |
| π€ AI Recommendations | Machine Learning, Django | Smart crop suggestions based on soil & terrain |
| Python, PostgreSQL | Automated risk assessment & farmer notifications |
# Example: SAR Data Processing Pipeline
def process_sentinel_data(sar_image):
vv_band = extract_polarization(sar_image, 'VV')
vh_band = extract_polarization(sar_image, 'VH')
waterlogging_index = calculate_wri(vv_band, vh_band)
hotspots = detect_anomalies(waterlogging_index, threshold=0.75)
return generate_alerts(hotspots)Impact: Helping Cambodian farmers prevent crop losses and optimize land use for food security πΎ
graph LR
A[π B.Sc. Data Science] --> B[π°οΈ Remote Sensing]
B --> C[π€ Machine Learning]
C --> D[πΎ Agricultural AI]
D --> E[π‘ Real-World Impact]
style A fill:#0ea5e9
style E fill:#10b981
Current Focus:
- π§ Deep Learning with TensorFlow & Keras
- π‘ Advanced Remote Sensing & SAR Data Processing
- π Time Series Forecasting & Statistical Modeling
- π Full-Stack Development for Data Applications
I'm always excited to collaborate on projects that combine data science with social impact, especially in:
- πΎ Agricultural Technology
- π°οΈ Remote Sensing & Geospatial Analysis
- π Predictive Modeling & Forecasting
- π°π Solutions for Southeast Asia
"From satellites 20,000km above to rice fields on the groundβtechnology should serve everyone."
πΎ Building AI for Cambodia's Future | π One Dataset at a Time