Dev Diaries

Texter Blue’s Team Dev Diary #7 – The development of TML – Texter Machine learning… with Vítor Moreira!

We move on with the seventh issue of our team Dev Diaries Series! Throughout these series, we will provide tips and tricks about the daily workflows, challenges, and learnings of a programmer/developer in a simple, disruptive and intimiste way! If this is the first one you read, don’t forget to check all our dev diaries!

For this Dev Diary number seven, we talked with Texter Blue’s Principal Consultant, and Alfresco Process and Content Services Certified Engineer, Vitor Moreira, about The development of TML – Texter Machine learning! Let’s check it…

What product are you responsible for developing?

Hello! I am responsible for supervising of the TML – Texter Machine Learning. It represents a whole new level of automatic extraction of information and data analysis that power the automation of key business processes not possible until now. Check it out here!

Why did the need for this product arose?

TML is like an AI-as-a-Service! It arises from the need to create a single (central) Artificial Intelligence data processing system. By other words, instead of having a client application connected to various artificial intelligence recognition systems, we have only one connection… into TML.

What challenges you had in the development?

One of the main challenges was related with the need to create a layer of communication that was common to various systems, and that allowed the recognition of faces, natural language processing, recognition of license plates, etc. In addition, and because AI systems are quite complex, we have had some challenges related to product scalability.

How do you solve those challenges?

Through the experience acquired during the development, we were able to implement a stable communication interface. To solve the performance problem, we’ve deconstructed everything into microservices, and with the help of Docker/Kubernetes we’ve been able to increase the number of instances of the services we need. In fact, we have reached the point of making this process automatic. In other words, the customer ends up spending only the computational power they need at that moment.

What impact the project had on the customer experience?

A major benefit of TML is related with GDPR compliance, and in the handle of personal data. TML can detect and anonymize Personally Identifiable Information (PII), either in pictures, texts or documents. But, If you want to learn more about all this process, please read my Technical Article, and this article.

So… we’ve got to the end of the seventh Dev Diary. Hope you liked! Leave your comments bellow, and If you want to learn more about us and our work, make sure you read our news and technical articles, and if you have any doubt contact us.