A little more than a year ago, OpenAI made ChatGPT available to the public. Since then, the world’s largest tech companies have raced to release AI assistants and politicians around the world have debated how to regulate AI without stifling innovation — all while journalists have breathlessly covered each new development in the space.
While the general public may see the current point in time as the golden age of AI, those with experience working with the technology know that these are still the early days. This has led to a more methodical approach to implementing the tech, one predicated on investigating how AI’s various applications can provide a real business benefit.
This is the line of thinking that guides GitLab, according to Taylor McCaslin, group manager of product for AI/ML.
“It’s easy to chase the latest and greatest tech, but it’s an entirely different challenge to ground its use in customer and company problems,” said McCaslin. “In many ways, the market is taking a technology and searching for problems it can solve.”
Rather than racing to be “first,” companies like GitLab are focused on getting AI “right” for both their customers and team. Continue reading to learn more about how GitLab and others are taking an intentional approach to AI along with how they’ve already implemented the tech into their daily work.
GitLab’s open-core, artificial intelligence-powered DevSecOps platform empowers 100,000+ organizations to deliver software faster and more efficiently.
How is GitLab embracing AI in its operations and making it an integral part of your company culture?
GitLab has been building AI features since early 2021, and our focus has always been on enriching the customer experience first. The initial couple of years were more about traditional AI forecasting models and building small, native, machine-learning models to drive operational insights. As generative AI has caught on and become more widely understood by non-technical folks, we’ve been evaluating how to leverage it as a company.
At GitLab, we build GitLab by using GitLab, so anything we build for customers directly improves our operations and is used in our daily work. We’ve built AI features into our product that help developers, security and operations experts, and non-technical users like product and project managers. These features help both GitLab and our customers develop software more efficiently and at a higher velocity.
In what ways does GitLab leverage its resources and expertise to implement AI in different areas of your work?
We’ve had a lot of fun over the past year working with excellent partners — including Google Vertex AI and Anthropic — to implement and scale our generative AI capabilities. We're working hand in hand with these AI vendors to push the limits of what Generative AI can do for our customers. It’s still in the early days, but the market is moving and evolving quickly. Some topics we’re digging into include how to effectively test AI models, mitigating hallucinations and prompting engineering optimizations as we see AI model context windows grow.
We’re also spending a lot of time looking at how our customers want to consume these features and technologies from different angles, including productivity, security and privacy. We ultimately are working to ensure we can confidently bring these technologies to our enterprise customers in responsible and compliant ways.
What are the benefits of incorporating AI into GitLab’s work, and are there any challenges or considerations that need to be addressed when implementing AI?
A big question is, how do you ensure very hyped technologies don’t distract from core business objectives? GitLab grounds its usage of AI on the problems and pain points of its customers and employees and uses AI when it makes sense. AI does come with a lot of complexity, and frankly, if you don’t have to take on that complexity, do the simplest thing possible.
GitLab grounds its usage of AI on the problems and pain points of its customers and employees.”
With that said, AI is definitely changing how we operate and build our product, and it comes with new challenges. It’s in the early days, but I’m excited to see where it goes. GitLab may not be the first to adopt some of these new technologies, but I’m certain we’ll adopt them appropriately as we find the right use cases that provide clear and direct value for our customers.
Multiverse helps people start careers in technology through its apprenticeship matching service and helps tech workers level up through its upskilling programs.
How is Multiverse embracing AI in its operations and making it an integral part of your company culture?
AI has the potential to revolutionize the way we work, to be an autonomous co-pilot to every worker. But what is vital to remember is that AI is pointless unless people can actually use it. We’re in the skills business, so we’ve thought seriously about how people in our business and working at our partner businesses will develop AI skills.
Multiverse has long been at the forefront of equipping people with the skills to develop AI tools. We already train more data analysts than the entirety of the United Kingdom university system, and almost 10,000 apprentices have been trained on Multiverse programs that include modules in machine learning and data mining.
But this year we went further and rolled out AI training to every single one of our apprentices and all of our employees. The ability to use AI has become an essential workplace skill, and we’re making sure everyone in our business and every apprentice we train is empowered with that vital skill.
In what ways does Multiverse leverage its resources and expertise to implement AI in different areas of your work?
Generative pre-trained transformers, or GPTs, are among the most accessible AI tools available today. The most commonly used manifestation is ChatGPT. Upon the opening of the ChatGPT API, we jumped at the opportunity to build MV-GPT, allowing us to make use of the tool while protecting our data.
Trialing it initially across smaller teams, we have now rolled it out fully and are seeing extraordinary results. Since its launch, our employees have asked more than 37,000 questions. As an organization of 850 people, this is already having a huge impact, saving hundreds of hours of time. Ultimately, we’re in the business of training and coaching — and we’re using AI to scale that.
AI can help ensure as many people as possible have access to great education, not by replacing human coaches and teachers, but by supporting and augmenting them. For example, AI can make it faster to mark assignments: When one of our coaches is augmented by AI, it takes them about a third of the time. This massively improves our efficacy while driving improved ROI for our customers — an AI-driven win-win!
What are the benefits of incorporating AI into Multiverse’s work, and are there any challenges or considerations that need to be addressed when implementing AI?
The key benefit of AI is its ability to turn scarcity into abundance. We can turn outstanding education, which has traditionally not been available to everyone, into a readily available and abundant resource. It means our people can be freed up to work on the most meaningful, exciting tasks rather than be burdened with administrative tasks.
What excites me the most about AI is the opportunity to create a generation of diverse future leaders.”
Of course, there are risks. When we’re using AI to make decisions or generate content, it has to be guided by human safety rails at every stage. That’s one of the reasons that training is so important, so people recognize these risks and mitigate them.
What excites me the most about AI is the opportunity to create a generation of diverse future leaders. AI is so new, and there is no pre-set supply of talent. There is a real opportunity for people from all backgrounds to become leaders in AI and other future technologies. It levels the playing field, allowing everyone to reach their full potential.
Point B is a consulting firm that specializes in business transformation and technology services.
How is Point B embracing AI in its operations and making it an integral part of your company culture?
We’ve embraced AI to improve and accelerate our internal processes and positively impact our customers. It extends our collective expertise, allowing us to combine our industry insights with transformative technology to create holistic solutions.
With years of experience developing AI solutions, including pioneering work in generative AI, we’ve gained insights directly from the hands-on creation and application of AI systems and software. We’ve identified several vital applications to enhance our operations in talent acquisition, resource management, knowledge management and data reporting.
AI extends our collective expertise, allowing us to combine our industry insights with transformative technology to create holistic solutions.”
We still view this moment as early in the story of AI, so we’re committed to a test-and-learn process, which is the best way for users to experience the technology and make it work for them. Just look at how ChatGPT has stormed onto the scene. Putting AI tools and capabilities directly in the hands of users goes a long way.
In what ways does Point B leverage its resources and expertise to implement AI in different areas of your work?
As consultants, we see nearly every business process and function and identify how they intersect and interact. Combining that point of view with our deep industry expertise gives us an advantage in determining how good any given AI system is or could be for our customers and our organization.
We are bullish on generative AI systems in particular, because of their ability to operate just like our people do and in ways they’re already familiar with, so integration is nearly seamless and the results are transformative.
What are the benefits of incorporating AI into Point B’s work, and are there any challenges or considerations that need to be addressed when implementing AI?
The opportunities are endless. For example, we’re seeing a rapidly growing list of use cases with the proliferation of generative AI — use cases that weren’t yet considered when working with other automation, like robotic process automation or other AI/ML systems.
One that has stood out from the crowd is Microsoft’s Copilot. It can offer point assistance to a myriad of tasks for knowledge workers throughout the day. In addition, more bespoke systems — some of which we are working on — will soon come online and offer AI that removes bottlenecks in processes and creates more efficiency across the board.
However, the primary challenge of AI is the learning curve of understanding the reality of data privacy and security. Ethical and strategic considerations will come into play as AI tools are integrated further into organizations. There’s also ambiguity around copyright law implications, if any, for generative AI models. Users must do their due diligence as these tools become more widely used.
