Step 1: Set clear goals.
Step 2: Determine pilot scope.
Step 3: Gather stakeholder input.
Step 4: Document current processes.
This week, I’m tackling Step 5: Evaluate technological readiness.
What, exactly, is meant by “technological readiness?” Assessing an organization’s technological readiness can make all the difference in success or frustration. Perhaps more importantly, it can steer your technological decisions. By “technological readiness,” I’m referring to:
- Infrastructure
- Computational Power
- Network Capabilities
- Knowledge and Expertise of Team
Regarding data infrastructure, do you have ways to collect data, store data and process data? For example, if you are collecting data from sensors, production lines, or customer feedback, you will need to have systems to collect and store large volumes of data. You will also need a way to determine if the data is of sufficient quality or not. If you choose to use Cloud Storage, you’ll have flexible infrastructure and you’ll be able to scale it as well. However, it might be difficult to get approvals for going “Cloud.” If that is the case, you might need to purchase a server to host the potentially large data sets yourself. How will you process the data? Will you want to process the data in real time? It’s possible to use third-party tools to help with data processing such as Apache Hadoop, Google BigQuery, Amazon Redshift, Microsoft Azure Synapse Analytics, or Databricks, just to name a few. (Note: Humaxa has absolutely zero affiliation with any of the tools above.)
Depending on how large your data sets will be, computational power needs will also be important to consider. It’s possible that you’ll want to train complex AI models as a part of your AI project. To start, however, I suggest that you focus on using cloud services instead. It’s simply MUCH easier to get started. If you do end up using your own hardware, you’ll need high performance compute resources like GPUs and TPUs. These can be VERY difficult to acquire, especially at the moment. You can bypass the global compute shortage by utilizing cloud services like AWS, Google Cloud, or Microsoft Azure. They can give you access to scalable computing resources and AI services that can be ramped up or down based on your needs.
Network Capabilities can be crucial to a project’s success. As data needs to be transferred between various parts of your technological infrastructure, you’ll want to make sure that the connectivity is fast and reliable. Testing for speed may be prudent if you’re not sure. If you might be running latency-sensitive applications, consider implementing edge computing solutions to process data closer to the source. (Edge computing solutions refer to the deployment of computing resources and services closer to the location where data is generated or where action needs to be taken, rather than relying solely on centralized data centers or cloud servers.) In addition to reducing latency, edge computer can improve response times and decrease bandwidth usage.
The knowledge and expertise of your team may be the last thing you consider regarding technological readiness, but in my experience, this can be the most critical and challenging item to ensure is in place. Does your team or do others you can ask to help have expertise in AI and Data Science? What is the level of their expertise? How will you know? If they have a basic level of understanding, can you provide training or upskilling opportunities? If you don’t have enough talent on your team (or on a team that you can pull from), will you be able to hire someone new? Or could you contract with a consultant to help you part time, and how will you know they have what you need? Sometimes new or existing team members will jump at the chance to learn about a hot, evolving field like AI and can learn how to interpret AI-driven designs or recommendations, without needing to actually do the work themselves.
If you would like to chat about these considerations or would like help with your AI experiment, please let us at Humaxa know. We’re always happy to help.
Carolyn Peer
CEO/Co-founder, Humaxa
