ML-Ops system for New York courier startup

Airpals

Team: Diana Mosquera, Francisco Gallegos

ml-oPSspatial intelligencelogistics

We partnered with Airpals, a New York-based courier startup, to develop a comprehensive data and AI infrastructure from the ground up. Our team built a robust data architecture that transformed their raw logistics data into actionable intelligence, significantly enhancing their operational efficiency and creating a competitive advantage in the fast-moving logistics sector. At the heart of our solution is an ML-Ops system that seamlessly integrates data collection, processing, and analysis. We implemented automated ETL pipelines to capture business data, creating a continuous stream of high-quality and diverse information. This foundation enabled us to deploy several machine learning models focused on spatial intelligence, forecasting and user segmentation.

The system's centerpiece is a comprehensive operations dashboard built with Panel (Python) and deployed on Google Cloud. This interactive visualization platform gives Airpals' team instant access to critical operational metrics and spatial analysis, allowing them to make data-driven decisions with confidence. The dashboard integrates predictive insights from our ML models with logistics and users data for a complete operational view.

The ML-Ops system provides Airpals with unprecedented visibility into its logistics network, enabling it to identify optimal hub locations, understand regional variations in performance and design loyalty strategies for its users. Beyond technical implementation, we established ethical data governance practices for Airpals' newly formed analytics team in the responsible implementation of AI. Our approach emphasized building sustainable data systems that respect privacy and deliver business value, going beyond mere data mining to create lasting intelligence capabilities that grow with their business. This project exemplifies our commitment to creating end-to-end data solutions that don't just solve immediate problems, but lay the foundation for continued innovation and growth in a data-driven world in a responsible way.