Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a customer analytics platform and in database ml in gcp
Data Engineer Data Scientist
I am a Data Engineer and Data Scientist specializing in designing robust AI applications and scalable data architectures. I combine scientific rigor with technical efficiency to transform complex prob...
About this Gig
I will design and build an end-to-end Customer Analytics 360° & In-Database Machine Learning solution using GCP and BigQuery ML.
What I offer:
- Churn Scoring & Real-Time API: Train attrition models in BigQuery ML (LOGISTIC_REG) and deploy FastAPI scoring endpoints for instant risk assessment.
- Customer Segmentation: Build an In-Database RFM Feature Store and apply K-Means clustering (VIP, At-Risk, Lost) for targeted marketing.
- Revenue Forecasting & Dashboards: 30-day sales predictions using BigQuery ML (ARIMA_PLUS) rendered on interactive Looker Studio dashboards.
Tech Stack: Google Cloud Platform, BigQuery ML, Python, FastAPI, Docker, Looker Studio, SQL.
Why work with me?
- In-Database ML: Eliminates costly data exports and leverages cloud speed.
- Production-Ready: Production-grade Python scripts, REST APIs, and clean SQL.
Please contact me before placing an order to discuss your data architecture!
Frameworks:
Scikit-learn
•
Google ML Kit
•
Panda
•
Other
Data type:
Other
Programming language:
Python
•
SQL
•
NoSQL
•
Scala
Tools:
MLflow
•
Other
APIs:
Google Cloud Vision API
•
Azure Face API
My Portfolio
FAQ
Why use BigQuery ML instead of traditional Python ML scripts?
BigQuery ML runs machine learning directly inside Google Cloud Platform without exporting data to external servers. This speeds up execution, avoids high data transfer costs, and scales easily with large datasets.
Do I need to grant you full access to my Google Cloud Platform project?
No. You can give me restricted BigQuery Admin and IAM Service Account permissions, or I can deliver clean SQL scripts, FastAPI code, and step-by-step setup guides for you to deploy on your end.
What data formats do I need to provide?
You can provide raw data from PostgreSQL, CSV files, Google Cloud Storage buckets, or direct BigQuery tables. I will handle the data cleaning, feature engineering, and Medallion pipeline setup.
