ELEMENTS TO CONSIDER IN AN ARTIFICIAL INTELLIGENCE PROJECT

In this text, we aim to provide some insight into how to approach an Artificial Intelligence project, outlining the different areas to consider and which of these should be included in your Work Breakdown Structure (WBS).

Infrastructure:

Before starting any project (whether artificial intelligence or otherwise), it's essential to understand the infrastructure you need, whether on-premises or in the cloud. Consider both costs and performance.

Model:

Although we often talk about LLM (Large Language Model), a somewhat overused acronym, when tackling an AI project, we might need a smaller model, such as SML (Small Language Model), which can be integrated into systems with fewer hardware requirements or smaller datasets.

Data:

The model will need to be fed with a dataset. Depending on the choices made beforehand, this data will need to be organized and "processed" (data governance) before it can be used.

Orchestrator:

'AI Orchestration' works by uniting the components of an Artificial Intelligence workflow. Automation, flows between different systems, and platform management itself are areas to be detailed.

Interface:

Another area to define is how users will interact with the system. This could be through a person, text, audio, etc., meaning the orchestrator will have to manage these requests along with the interface.

This is a brief overview of high-level elements to consider when organizing an Artificial Intelligence project.

I'm Daniel Cabezas from AvicProjects, applying the latest technologies within your company to create value. Feel free to contact me and we can talk.