The Client

Metrinomics provides tools and services to improve Customer Success and to engage customers, as well as the mathematics to drive adjustments all along the interaction journey. The focus is on B2B markets with rich complexity and high impact on customer relations.

Metrinomics is based in Berlin and looks back to a history of more than 20 years in using AI to create smart algorithms that reflect subjective drivers of businesses. They have large international tech companies and financial institutions as their clients.

Jungsoft helped Metrinomics to modernize their technology stack, implement agile processes, and collaborated on the development of both internal and client-facing systems.

The Problem

Metrinomics has been in the market for more than 2 decades. While they accumulated (a lot of!) experience in their area, their software systems also accumulated some technical debt and started to get in the way of their business.

We had two main challenges: build new products that were modern and compelling, and improve the existing systems and processes.

The Solution

To reach our goal, we realized that we would need to work on multiple fronts. Using modern technologies was only part of the solution, we also needed to work on implementing agile processes and ultimately lead to a business transformation.

The Implementation

Our collaboration involved a lot of on-site time, which was vital for understanding the culture, processes and workflows that were part of the company's day-to-day.

After a thorough analysis of the existing systems and considering the roadmap, one of the first things we did was a migration to the cloud. By combining technologies such as Docker, Gitlab CI/CD and AWS Cloud, we were able to see a drastic increase in the frequency of new releases and decreased maintenance costs, as well as mitigate concerns regarding the systems' scalability.

We also worked closely with their in-house engineering team to improve the performance of legacy systems, having achieved more than tenfold improvements in some cases.

Another big part of our work revolved around improving and building new data pipelines for their big-data systems, where we introduced the use of Tensorflow with Google Cloud and built a new system for easily configuring and managing different AI algorithms.

Our development squad also built two new customer-facing systems, that integrate with the existing data processing pipelines but are focused on a new unexplored market. These systems were built using modern technologies such as React, React Native, Python, Elixir, GraphQL and PWA.

The Results

During our collaboration, we worked very closely with Metrinomic's in-house team, and seeing improvements across the whole organization was a very rewarding experience for us.

While improving the legacy systems enhanced the experience of long-term existing clients, the new systems attracted new users from multiple countries around the world and have been running successfully in production for months.

Technology Stack

Elixir

Phoenix

GraphQL

React

Apollo

Material-UI

Bootstrap

PostgreSQL

MongoDB

Python

TensorFlow

Docker

Kubernetes

AWS Cloud

GitLab