Project Detail

In-House ML Deployment Framework Development

Developed and improved an internal Python framework used across departments to standardize the deployment and operation of business-critical machine learning applications.

Role
Software Developer
Context
Telecommunications
Domain
ML Platforms
Duration
5 months
Status
Finished

What changed

  1. Integrated automated monitoring capabilities with Databricks, removing manual configuration steps for internal teams.

  2. Refactored framework components to improve maintainability, readability, and overall developer experience.

  3. Translated requirements from developers, internal users, and stakeholders into framework capabilities that supported adoption across departments.

Technology

Python

13 technologies

TensorFlow Pydantic scikit-learn pandas PySpark SQLAlchemy CausalML Databricks Connect PyYAML uv pytest mypy Black

SQL

3 technologies

PL/SQL Spark SQL SQLite

Databricks

6 technologies

Databricks Runtime Unity Catalog Databricks Clusters SQL Warehouses Databricks Asset Bundles MLflow

CI/CD Pipelines

1 technology

Azure DevOps

Azure

5 technologies

Azure Container Apps Azure Container Registry Azure Virtual Machine Scale Sets Azure Key Vault Azure Storage Accounts

Signals

machine-learning mlops platform-engineering devops cloud-engineering internal-tooling