The 3 Biggest Ways AI Is Helping IT Professionals

AI is Transforming the Role of IT Professionals

For decades, IT professionals have been bombarded with unfulfilled promises about technology solutions. We’ve seen seemingly endless waves of tools that promise to increase reuse, accelerate project delivery, and empower every worker to perform as effectively as the best employee. Unfortunately, most of these tools have fallen short.

The good news? AI is changing that. High-performing conversational interfaces powered by AI are helping IT professionals with tasks such as improving time to produce quality work products, enhancing work transferability, and navigating massive knowledge repositories. This couldn’t have come at a better time: The demand for IT professionals has always exceeded the supply, and now we face the added challenge of a volatile economic climate that’s causing budgets and teams to shrink. It’s more critical than ever for IT to operate at maximum efficiency.

In this article, we’ll look at three ways that AI can help as well as some considerations for adopting it into your organisation’s IT.

Accelerating Developer Workflows

AI coding assistant tools like GitHub’s Copilot, Amazon’s Codewhisperer, and Replit are able to provide a “pair programming” experience for developers with capabilities ranging from turning natural language prompts into coding suggestions across dozens of languages to providing code reviews and suggestions for improvements. This in-flow support can dramatically reduce context switching and cognitive load for developers, compressing the time it takes to create high-quality work products.

Before the advent of these tools, code handover was almost always an extremely time-consuming and arduous knowledge transfer task, especially when different time zones were involved. I’ve experienced this firsthand: In the early 2000s, I was involved in a handover project across time zones that was projected to take three to four weeks and ended up taking the better part of a year. (Talk about an unfulfilled promise.)

With AI coding assistants, both sending and receiving individuals now have a powerful assistant to help document previously undocumented code bases (a perennial problem) and summarise logic in plain language, including references to technologies with which the receiving individual is more familiar. In fact, a colleague who works in numerous programming languages and who has become skilled at learning new languages over time, recently remarked that having an AI assistant has helped to understand new languages at a far more rapid pace. The reason is their ability to explain how two languages are both different and similar in the context of a particular project.

Increasing Knowledge Base Access

AI is also changing the game when it comes to knowledge base access for IT professionals. One of the hardest parts of finding documentation when stuck on a specific task is taking the specific problem, translating it into verbiage that the vendor would have used, and then searching for and identifying the correct document to address it.

Organisations invest a tremendous amount of resources into creating documentation, but the ability to use that documentation has historically been very difficult.

I recently chatted with an executive whose company took its internal documentation—plus Google’s publicly available documentation—and indexed it using Google’s Vertex AI to create a conversational help search engine for its consulting team. With this option, instead of hunting down the documentation, the team can pose a question detailing the problem and be given a summary of the solution plus a link to the original documentation.

The Long And Short Of It: AI Is Helping With Generalization

In a broader sense, the examples that I have shared here demonstrate an important goal of AI researchers and developers: generalisation. Code commenting, pair programming, and search are all examples of more general use cases like summarization, search, and text generation. For example, at my organisation, a user researcher is using OpenAI’s ChatGPT to help synthesise bullet-point notes from transcripts from multiple user interviews into research summaries, which has saved hours of editing and collation and increased accessibility to her work by the rest of the organisation.

The demand for these types of natural language interactions with systems will only grow as generative AI tools continue to develop. This creates a predicament for IT teams: On one hand, these tools are helping them be more productive. On the other hand, they are now being asked to provide these tools in a specific context for their business, and not all IT teams are prepared to do that.

Getting The Most From AI

AI is making the jobs of IT professionals easier, but how can they ensure they’re getting the most from the technology?

One thing that IT professionals—and anyone who uses AI—should continually be working on is refining their prompt engineering skills. Successfully using large language models like ChatGPT or Google Bard depends heavily on giving the programs the right inputs, i.e., asking the right questions.

It’s easier said than done. This is a field that has been in the mainstream for approximately twelve months. There aren’t any robust programs, nor is there a standard curriculum, for prompt engineering training

FAQ