I am a Data and ML Engineer based in Paris. I spend most of my time building machine learning systems that actually work in production not just in notebooks.
My background is a mix of IoT sensor data, predictive analytics, and cloud deployment. I have worked on everything from real-time anomaly detection on industrial hardware to customer churn models running on Azure. I like the full pipeline: raw messy data in, deployed API out.
Currently open to Data Engineer / ML / Devops / Mlops roles.
📍 Paris, France
📬 abbasi-anees.ahmad@outlook.com · Linkedin
| Tools | |
|---|---|
| ML & Deep Learning | PyTorch, TensorFlow/Keras, Scikit-learn, XGBoost, LSTM, YOLO |
| Computer Vision | OpenCV, DETR, CNN, Transfer Learning |
| Cloud & MLOps | Azure ML, Docker, MLflow, Azure Functions, CI/CD |
| Databases | InfluxDB, MongoDB, Azure Blob, Event Hub, SQL |
| Data Engineering | apache Spark, Kafka, Airflow, Docker, Kubernetes |
| Interfaces & APIs | FastAPI, Streamlit, Power BI |
| Certifications | DP-100, AZ-400, AZ-104, PL-900, DP-700 (Microsoft), Databricks Academy: Lakeflow Jobs, Lakeflow Connect, Spark Declarative Pipelines, DevOps for Data Engineering |
⚪ End-to-End-Retail-Data-Pipeline-on-Databricks
End-to-end retail data pipeline on Databricks using PySpark and Delta Lake, built with Bronze–Silver–Gold architecture and connected to Power BI for analytics.
🔴 Real-Time Churn Prediction — Azure
Event-driven pipeline that scores customer churn probability as events happen.
Built on Azure Event Hub → Functions → ML endpoint → SQL, with a Power BI
dashboard on top. The goal was persistent, real-time risk tracking — not a
batch job that runs overnight.
View Repo
🟠 Turbofan Engine Failure Prediction — LSTM
Predictive maintenance model on the NASA C-MAPSS dataset. Framed as a sequence
learning problem — take multi-sensor readings, predict how many cycles until
failure. Achieved R² = 0.87, deployed as a FastAPI + Streamlit prototype so
engineers can query it directly.
View Repo
🟡 IoT Anomaly Detection Pipeline
Built during my time at Artifeel — real-time telemetry processing from IoT
devices, anomaly detection using Scikit-learn, deployed on Azure ML. The tricky
part was handling irregular sensor signals and temporal degradation patterns
without drowning in false positives.
- Full-time Data & ML roles
- Freelance ML / Data Science projects
- Computer Vision & IoT Analytics consulting