Work Smarter: How Sitation Product Support Applies AI Tools

Sitation is known for its AI-powered digital transformation services and software. When it comes to supporting our products, we harness the power of AI tools to streamline our processes. Sitation Product Support uses AI tools in a variety of ways to efficiently get answers and identify solutions for our clients.

AI: Tools for Humans

It’s important to note that I view AI as a tool for humans, not a replacement for humans in the support process. AI helps us to locate information quickly, to eliminate much of the “legwork” and analysis that may have been done manually or with human-programmed processes previously. AI tools can also help to prove (or disprove) theories about what may be causing specific issues, so I can more accurately frame and convey the factors contributing to issues that our Product Development team may need to address.

I have found AI to be one of the best research tools available. For example, our Product Support team uses AI tools to provide facts and context about systems that Sitation doesn’t control. Whether identifying configuration details in third-party software, gaining insight into multi-platform data transformation scenarios, or explaining why data may be handled in a certain way in one system compared to another, AI is great at locating and summarizing complex data. This allows our team to spend less time researching and more time doing what we love: facilitating fixes and communicating with clients.

“Why is this broken?” : AI for Code and Data Analysis

One of the most useful and educational applications for AI in Product Support has been analyzing Sitation’s own code bases. AI has been incredibly useful to me, especially when working with large code bases, in explaining how disparate aspects of code work together to move and update data within databases. On a personal professional-development note, this is helping to improve my understanding of code generally, as well as my comprehension of how specific functions operate.

I run AI code base analyses to identify:

• Potential bugs
• Where changes would need to be made if requested features were to be implemented
• Processes in which database records are created and updated

Another common use case is applying AI to analyze specific database records with a given code base as context. I have found this approach useful for needs like finding problematic field values in a given context, such as whether a field contains an array of values when only a single text string is expected by an application. Similarly, I have used AI for identifying relationships and dependencies between collections in a database, so it’s clear how a change in one record may impact another. This represents a major process improvement that frequently saves our Product Development team the time of having to answer many of the support-related questions that AI can now help to answer, which allows them more time to focus on building cool stuff.

“Write that down.” : Documentation Generation

Along the way, I help to capture and document the insight that AI has provided. I have used AI to generate article drafts for future reference, to search and summarize existing documentation, and to recommend updates for potentially obsolete content. Humans like me still need to review and validate these recommendations, since they are not always initially accurate; even with increasingly better models, AI hallucinations can still happen! But the time savings continue to increase, making our Product Support team more efficient in helping to resolve and document resolutions for technical issues for our clients.

The Way Forward

Here in Product Support, like the rest of Sitation, we are constantly assessing new ways to apply AI tools to innovate and expedite processes, so we can continue to deliver world-class experiences for our clients and to help them win on the digital shelf. It’s clear to me that creative and responsible application of AI tools is the way forward, and I look forward to continuing to help innovate our processes and technology. Thanks for reading this! (Believe it or not, no AI tools were used in the writing of this blog post.)