A total of 36 submissions, 4 cases that are actually implemented with clients, and 1 grand winner: that’s the outcome of the Azure OpenAI Hackathon at Macaw. Since the early developments in Artificial Intelligence (AI), Macaw has been at the forefront of acquiring knowledge and applying it to our clients. Thanks to the results of the Hackathon, these developments and opportunities can accelerate even more.
Out of the 26 submitted cases by enthusiastic Macaw employees, the Hackathon jury selected 4 that best align with clients or could serve as a great source of inspiration for them. All submitted cases were created as demos within six to seven hours and presented during the Hackathon. “Within two weeks, you could further develop such a demo into a complete tool,” says Macaw’s Head of Technology, Michel Heijman. “The speed at which we can create the demo demonstrates how far we have come with OpenAI and how quickly you can build something decent with it.”
Power Platform CoE Copilot
The first case highlighted by the jury was an Azure OpenAI-based Power Platform CoE Copilot, built for a client in the food sector. Macaw already provides the Centre of Excellence (CoE) for Microsoft Power Platform to that specific client, encouraging citizen developers to build apps. “This naturally raises a lot of questions for our employees,” explains Michel. For example, questions like: how should I build a Power App according to the company’s standards? Or: what similar solution already exists within our company? “The Power Platform CoE Copilot has been deployed as a kind of solo service center for all these questions, both technically and organizationally. We find that this takes a lot of work off our teams’ hands and makes the client’s citizen developers more self-sufficient.”
For a leasing company, Macaw built an Azure OpenAI-powered Content Writer that assists the client in writing and updating texts for the catalog site, where lease drivers configure their cars. Once the car is fully customized, a summary text appears, indicating, for example the chosen type of wheels, brand of the car, type of car radio, and tires. Each combination requires a separate text that matches the brand’s tone of voice and the chosen components, catering to the needs of the lease driver. The Content Writer autonomously generates and updates all these texts.
Customer Experience Copilot
We’ve all experienced it: you visit a website with a specific question, only to find a vast amount of information and related answers, but not the one you’re looking for. Frustrated, you give up and try your luck with a competitor. This risk particularly exists with large companies that have extensive content. For the Macaw Hackathon, an AI copilot was designed to scrape all the content from a site upon request and provide highly detailed answers to questions.
Michel says: “We regularly showcase the copilot to clients in a demo, and people are consistently amazed at what AI is already capable of. It makes clients think about other ways they could use this tool. It also provides an opportunity to discuss topics such as data and Digital Experience Platforms, or how to use the copilot to shape the entire customer journey.” This marketing-inspired tool has broad potential. During the Hackathon, the tool was further developed: the copilot can now provide not only written information, but also determine when it’s more appropriate to offer visual answers (such as in a graph or bar chart).
Smart Talent Planner
The fourth chosen OpenAI application from the Hackathon emerged as the winner. This tool is particularly suitable for large companies where employees change positions every few years, either due to reorganizations or policy changes. But the Smart Talent Planner also offers a solution for putting together project teams. Reassigning people or setting up teams is usually a complex task, and putting the right employees in the right places is a real challenge.
The Smart Talent Planner is a bot that is trained to match all employees of a company, including international employees, to a particular position based on keywords in resumes. In doing so, the bot takes into account experience, skills, knowledge and the possibility of filling a position based on working remotely. The Smart Talent Planner chooses the most suitable candidates, regardless of location or part of the company.
Macaw: Microsoft Preferred AI partner
As mentioned, some of these cases have been available to Macaw customers for some time, and continue to evolve. ‘Macaw has been at the forefront when it comes to innovative technology for years, and so it is now on a roll with OpenAI. We are getting our employees ready to start working with it, and in the meantime we have our own examples and demos,” Michel explains. Macaw has been named one of the Preferred AI partners in the Netherlands and Germany by Microsoft for good reason. Macaw has over 25 years of expertise in the Microsoft stack. Does your company also want to get started with AI? Make an appointment with one of our consultants and get inspired. Contact us here.