Methodology

We are designing, recommending, architecting and building solutions in various sectors either for business, government or NGOs looking to harness the power of Machine Learning into their day to day activities as well as strategic decision support.

During 25 years of experience in software engineering, infrastructure setup and deployment sophisticated solutions, we developed our own approach and principles to build solutions and tools.

Start from a problem towards solution

Although this looks natural or expected, you will be surprised to know that it is not uncommon mistake to start proposing solutions that turns to be irrelevant or not efficient due to lack of understanding of the original problem. In this phase we are focusing on understanding the problem in the deepest way possible before jumping into recommending or designing solutions.

Start with a set of alternative  solutions

It is not very hard to offer a set of solutions for any problem, finding the fitting one was always the challenge.  Selecting the best solution is done through various means like challenging alternatives by different architects and design after evaluating different approaches using comparative studies offered by the science community.

Design for scalability

In many situations, operational and scalability challenges might be larger than developing the solution itself. Because of the nature of machine learning, operational complexity, performance efficiency as well as cost efficacy all should be considered before writing the first line of code.

Never reinvent the wheel

Some of the KPIs of our engineers is the time to production, cost per line and lines of code per function. That’s why it is very common to find integration with already available AI providers while our solution is orchestrating the process and adding the unique value for our customers, this is not only reducing the total cost of ownership but reduces the time to market dramatically.

Sounds interesting?