How to use Midokura's Technology Radar

Introduction

Technology is advancing rapidly, with new technologies and innovations constantly emerging.

It is important to actively seek out innovations and new technologies and periodically question established technologies and methods.

But, it is also important to wisely choose which technologies to use in our daily work and in the different projects we are carrying out. As we all know: There is no silver bullet.

What is the Midokura's Technology Radar?

The Tech Radar provides an overview of different technologies, including languages, frameworks, tools, and patterns, as well as platforms, that we consider 'new or noteworthy.' The radar does not cover all established technologies; instead, it focuses on items that have recently gained significance or undergone changes. Items previously featured in the radar are not listed on the homepage but remain available in the complete overview and search.

How it is created

This Technology Radar is an initiative from Midokura to share information on what we see valuable to watch, try or adopt. It is currently maintained by the Advanced Development department in TPSO, together with our CTO Dan Dumitriu. We aim for it to become the seed of an AITRIOS-wide technology radar with SSS responsibles for the different quadrant.

How should it be used

The radar serves as an overview of technologies that we believe everyone in the teams should be aware of at present.

Its goal is to guide and inspire daily work within the teams. Additionally, it aims to provide valuable information and a high-level perspective to enable decisions to be made with a deeper understanding of the subject matter, resulting in more informed and coordinated choices.

We categorize the items into four quadrants, and sometimes, when it's not entirely clear where an item belongs, we choose the best fit.

The quadrants are:

Agentic AI Infrastructure This quadrant focuses on infrastructure, orchestration, memory, and systems that enable autonomous AI behavior across cloud and edge.
Data Center & Cloud AI This quadrant covers core AI infrastructure at scale, including GPU/HPC, storage, networking, and secure cloud AI services.
Edge & Physical AI This quadrant covers physical AI at the edge, including robotics, sensors, embedded inference, real-time control, and autonomous deployed systems.
Hybrid Infrastructure & Operations This quadrant captures cross-tier operations, connectivity, observability, and hybrid management spanning data center and edge.

Each of the items is classified in one of these rings:

Adopt: We wholeheartedly recommend this technology. It has been extensively used in many teams for an extended period, proving its stability and utility.
Trial: We have successfully implemented this technology and suggest taking a closer look at it in this category. The aim here is to scrutinize these items more closely with the intention of elevating them to the 'Adopt' level.
Assess: We have experimented with this technology and find it promising. We recommend exploring these items when you encounter a specific need for the technology in your project.
Hold: This category is somewhat unique. Unlike the others, it advises discontinuing or refraining from using certain technologies. This does not necessarily imply that they are inherently bad; it often may be acceptable to use them in existing projects. However, we move items here when we believe they should no longer be employed, as we have identified better options or alternatives.

Contributing to the Midokura Technology Radar

Contributions and source code of the Midokura Tech Radar are on GitHub: Midokura Tech Radar on GitHub (private)