Martin Sommerseth is the Managing Director of Visma Resolve. Before taking on a management role, he worked in development and studied industrial economics and technology management at the Norwegian University of Science and Technology.
Whenever Martin isn’t at the office – strategising and sharing insights – you’ll probably find him outside doing downhill or randonee skiing. Read on for his journey within the world of AI, and how his passions got him to where he is today.
Hi, Martin! What initially got you into the world of AI?
“I initially found this direction interesting during my study programme. Being able to simplify tasks that are normally time consuming caught my attention during my first years of studying.
Then, my interest in AI increased a lot during my summer internship at Visma in 2017. I worked with using AI to solve timetables for schools, ultimately enabling teachers and students to find better timetables compared to solving it manually.
After my internship, I was fortunate enough to spearhead the establishment of Resolve, which was originally an AI unit within Visma, along with a team of four talented people. Within just a few weeks of embarking on my professional journey, it became abundantly clear that I had found the perfect place for me.”
“Improving people’s lives is what drives me the most. My passion is to make software that helps others make the best decisions, as quickly as possible, to solve their problems. To me, the potential that AI has to facilitate this is undoubtedly the most captivating aspect of it.”
Martin Sommerseth
How would you describe the company culture of Visma Resolve?
“Young, fast paced, and hungry for challenges. We have a high level of autonomy and trust.”
What does your typical workday look like?
“For me, a typical workday includes talking to existing and new partners to maintain a good connection, and to ensure that our goals are aligned with them.
Not a day goes by without looking at Resolve’s objectives and key results. And, together with the team, we strive to make the best strategies to reach our goals. I also dedicate time to provide Resolve’s product teams with the insight they need to autonomously make the best priorities and decisions.”
What do you think the current AI revolution means for the future?
“I think that more workplaces will be replaced by software, such as retail and customer service, and manufacturing.
For our customers, I believe we’ll see a huge trend in AI and machine learning automating their data entry, document processing, and administrative tasks – which, ultimately, frees more time for them to spend on core business.
The AI revolution will be a significant contribution to people getting to spend more time on what matters the most to them, both in terms of their work lives and private lives.”
What are some of the most challenging AI projects you’ve worked on?
“The most challenging AI product I’ve worked with first-hand is without a doubt our route planning product. It aims to solve the operational planning problem for at-home health care units in Europe. More specifically, we provide a platform for the home care units’ nurses, with patient tasks in a well-planned order.
It’s been challenging to make this a user-friendly software, as there are so many human aspects to take into account when planning the routes. That’s why this has been such an interesting process. It has enabled users of the software – the nurses – to spend up to 14% more time with their patients, compared to the previous routes they worked with.”
What’s the main thing working with AI has taught you?
“Working with AI has taught me the importance of user friendliness, and the need for AI software that users trust and feel in control of. If you aren’t able to build trust, your product won’t be sought after.”
What do you think are some of the ethical concerns and considerations that arise when working with AI, and how should they be addressed?
“There are numerous ethical considerations that must be taken into account when developing and offering AI software. Some key considerations include:
- Transparency: It’s crucial to ensure that users understand the reasoning behind the AI’s recommendations. Clear communication and explanation mechanisms should be implemented to promote transparency and foster user trust.
- Fairness: AI systems should be designed to treat individuals fairly and avoid biases. Careful attention must be given to data selection, algorithm design, and model training to mitigate potential discrimination.
- Accountability: Establishing clear lines of accountability is essential to address the decisions made by AI systems. Defining roles and responsibilities helps ensure that the impact of AI-generated decisions can be properly attributed and addressed, if necessary.
As more and more people experience the power of AI, I think they’ll treat it as a tool that helps them make better decisions.
Look at the healthcare sector as an example. Seeing the power of generative AI, large language models, and ChatGPT leaves no doubt about this technology playing an important role in diagnosing patients. Doctors will always have the responsibility, but with new tools that let them efficiently and precisely provide feedback and diagnose patients.”
“Fostering a mindset of openness to experimentation, and viewing AI as a tool in an extensive toolkit, helps to drive progress. By embracing AI as a complementary tool, we can leverage its potential while simultaneously upholding ethical principles.”
Martin Sommerseth
How does Visma Resolve push the boundaries of innovation with AI technology?
“We’re continuously looking for opportunities and products that don’t already exist within Visma. We are at the forefront of AI technology, and we always aim to identify the most valuable problems to be solved – and we will solve them.
In the last five years, we’ve built a team of almost 30 people that have become experts on a technology that suddenly almost every SaaS company in the world would like to have.”
What would you say to someone who’s looking to get into the field of AI?
“It does, of course, depend on what you wish to achieve. Are you mainly doing it because you want the AI competence, or are you hungry for creating valuable products? I always advise others to start with identifying a valuable problem. What is it that you want to solve?
Ask yourself what you think are the biggest pains in your life, or in your customers’ lives, and start from there. Then, why not ask ChatGPT about how you can gain the right knowledge and skills to solve this problem? ChatGPT is much better at giving efficient and precise advice than I am. In fact, it’s better than the advice that most others would give, as well.”