Data Engineering | Data Science & Machine Learning | Cloud & DevOps | GIS & Remote Sensing | Project Management

I build end-to-end data solutions that turn raw data into actionable insights—combining geospatial intelligence, cloud platforms, artificial Intelligence, and machine learning.

Languages:
Python | SQL | R | Scala | Bash

Data Engineering & Architecture:
ETL/ELT | Data Lakehouse | Data Warehouse | Medallion Architecture | Data Modeling | Batch & Streaming Pipelines

Big Data & Processing:
Apache Spark | PySpark | Apache Kafka | Apache Flink | Airflow | dbt | REST APIs

Data Platforms & Storage:
Databricks | Delta Lake | Snowflake | PostgreSQL | MySQL | Elasticsearch | Parquet

Cloud:
AWS | Microsoft Azure | GCP

Machine Learning & MLOps:
Scikit-learn | TensorFlow | PyTorch | MLflow | Isolation Forest | Random Forest | Prophet

DevOps & Observability:
Docker | Kubernetes | Git | GitHub | Prometheus | Grafana

GIS & Remote Sensing:
ArcGIS Pro | QGIS | GeoPandas | Rasterio | Sentinel-2

Visualization:
Streamlit | Power BI | Tableau | ArcGIS Dashboards | Kibana | Grafana

• Designed and implemented an Intelligent Power Monitoring & Distribution System for Kubernetes infrastructure, integrating real-time power and system telemetry into a centralized monitoring platform.
• Built Python-based data pipelines to collect and process power, node, and pod metrics every 15 seconds using Prometheus, Elasticsearch, Kibana, and Grafana.
• Developed Isolation Forest anomaly detection and Prophet forecasting workflows to identify abnormal operating conditions and predict power consumption using MAE, RMSE, and MAPE for evaluation.
• Containerized and deployed a KubeRAG MLOps monitoring application using FastAPI, Docker, Kubernetes (k3s), ChromaDB, Ollama, MLflow, and Git/GitHub.

🔗 https://github.com/cnero101/Intelligent-Power-Monitoring-and-Distribution-System-IPMDS

• Built an end-to-end real-time pipeline leak detection system using Databricks and Apache Kafka.
• Implemented a Bronze, Silver, and Gold Medallion Architecture for scalable streaming data processing.
• Automated HTML email alerts through Azure Logic Apps and Gmail when anomalies or critical pipeline conditions were detected.
• Automated critical leak detection and email alerts, enabling notifications within 30 minutes of detection.

🔗 https://github.com/cnero101/Real-Time-Pipeline-Risk-Monitoring-with-Databricks

• Built an end-to-end real-time pipeline risk monitoring system using Azure Event Hubs, Functions, and Data Lake Storage Gen2.
• Streamed and processed sensor telemetry from 5 simulated Alberta pipelines, using a Random Forest model for anomaly and risk detection.
• Applied Isolation Forest for real-time anomaly detection, with model tracking and management using MLflow.
• Automated HTML email alerts through Azure Logic Apps and Gmail when anomalies or critical pipeline conditions were detected.
• Developed a live Streamlit dashboard for pipeline status, risk distribution, and sensor trends, with 30-second automatic refresh.

🔗 https://github.com/cnero101/Real-Time-Pipeline-Risk-Monitoring-with-Azure

• Designed AWS streaming pipeline (Kinesis, Lambda, S3, SNS)
• Processed structured and semi-structured data into Parquet format
• Built Streamlit dashboard for real-time anomaly detection and monitoring

🔗 https://github.com/cnero101/AWS-Data-Engineering-Project

• Mapped burned area extent and severity from the 2024 Jasper Wildfire Complex using Sentinel-2 satellite imagery
• Compared four ML/DL models — Logistic Regression, Random Forest, MLP, and CNN — for pixel-level burn classification
• Best model (MLP) achieved F1 = 0.985 and Cohen's Kappa = 0.984 on a spatially-split, leak-free test set
• Detected 516.7 km² of burned area using NDVI, NBR, and dNBR spectral indices

🔗 https://github.com/cnero101/Wildfire-Severity-Assessment-in-Alberta

• Built spatial analysis workflows using Alberta wildfire data (2006–2024)
• Applied clustering (KMeans + PCA) and logistic regression for risk prediction
• Developed interactive dashboards (Streamlit + Jupyter) with hotspot maps and KPIs

🔗 https://github.com/cnero101/alberta-wildfire-analysis

• Performed GIS-based multi-criteria analysis for optimal facility placement
• Integrated demographic and spatial datasets for decision support

• Build scalable ELT/ETL data pipelines (batch + streaming)
• Design data lakehouse and cloud data architectures
• Develop machine learning models to solve real-world problems
• Perform geospatial and statistical analysis
• Build interactive dashboards and monitoring solutions for decision-making

I’m open to working on:

• GIS & spatial analytics projects
• Data engineering pipelines
• Machine learning applications

• 💼 LinkedIn: https://linkedin.com/in/ifeanyi-e-njoku
• 📧 Email: ifeanyinjoku2@gmail.com
• 🌐 Portfolio: https://ifeanyinjoku.com