Ilan B.

As Lead Tech & Data, my role encompasses both the technical aspects and the tech and data. You led the development of the backend of the application using technologies such as NodeJS, Express and MongoDB. My expertise extends to designing a backend schema architecture to optimize workflow and business logic. I ensure efficient synchronization between the backend and the No-code front-end made with Bubble.io.

An essential part of my work is to ensure optimal backend performance by implementing micro-services with Cloud Functions and Pub/Sub, or Cloud Run for deployment. I also worked on the integration of partners via their API.

For the Data part, I set up integration processes with Fivetran, created extraction and transformation layers (DBT) and used Lookerstudio to develop relevant dashboards.

Regarding internal processes, I fostered smooth communication using Slack and Notion. I manage agile development cycles (Sprint), we document technical advances a lot and perform PR reviews on Github to maintain code quality.

As Lead Tech & Data, I played a central role in connecting the technical and data aspects, optimizing performance and fostering effective collaboration within Qeeps.

I joined Flink after the acquisition of Cajoo as Lead Data Scientist to work on Last Mile projects.

Design the forecast demand to create the planning of runners and preparers in the Hub.

The forecast demand is linked to many projects (for example, order forecasts, no-show forecasts, productivity of our runners and preparers, etc.

- Order forecast: We use different Machine Learning models to forecast the number of orders in the next 4 weeks.

- No-show forecast: We are working on forecasting the percentage of absences in order to adjust the number of runners and preparers.

Currently Data Scientist at Cajoo on the Delivery scope.

I have worked in recent months on various projects:

- Creation of a webapp to automate the creation of schedules for delivery drivers with the use of Machine Learning algorithm to predict the need for delivery drivers based on the order history.

- Optimization of delivery driver schedules, running all the needs with a limited number of delivery drivers, and many other constraints.

- Cross team project with the Software Engineer team to create a scoring system to designate the best delivery driver for each order.

- Use of dbt to retrieve raw production data and transform the data so that it is easily accessible for business teams.

- Dashboarding to track activity in real-time with Metabase.