What does it take to productise an AI solution? Lessons from two development projects.
05 | 2025 Olli Laine, Kipinä Co-founder & CEO
An AI solution often sounds like it involves a huge amount of maths, black boxes and consultant jargon. But when you look at the practical level, building an AI product is ultimately not so mysterious. It requires the same as any successful product: a clear problem, a bold vision and a good team.
In this blog, I'll open up about two very different Kipinä customer cases - and what we learned with them about building an AI product. The first story starts with the everyday life of a doctor. The second starts from a global goal to make ESG measurable.
1. RIDA – A solution born out of a doctor's frustration
Ridha Bennour is not a coder. He is a doctor who became frustrated with constant typing.
One remote appointment – same patient again, same notes again. It felt like I spent more time staring at a screen than meeting people.
And then an idea came to mind: what if artificial intelligence could handle those entries?
This is how RIDA got started. The spark came when the idea had to be turned into a real application—one that didn't require integration or training, but was still reliable. Doctors had to be able to use it right away.
The goal was ambitious: a tenfold increase in the efficiency of doctors' notes.
In practice, this meant, among other things, the use of model sentences and a super-simple user interface.
Three things stood out in the project:
Collaboration: Rida Software recognized the problem. We knew how to build the solution.
Quick benefits: It is not always necessary to use artificial intelligence for eco-friendly steps.
Trust: Since there was no previous development team, we were more than just implementers—we were partners.
Ease of implementation: "Plug'n play" is not just a buzzword – RIDA really works without integrations.
RIDA is now used in several areas of well-being. And yes – it really does free up time for people. The potential for further development is enormous!
CASE RIDA
"Kipinä gave us so much more than just a development team. Through our collaboration, the Kipinä ecosystem has provided RIDA with visibility and networks that have also boosted the business start-up itself. The fact that we finished the RIDA app on time and on budget is a testament to our collaboration!"
Ridha Bennour
Gargantuan Medical Solutions
2. Planet AI – When responsibility needs to be measurable
The second story begins in a completely different world.
Planet Company had developed an idea of how companies could truly demonstrate their responsibility. ESG was full of great goals—but where was the concrete action? Planet AI woke up and an idea was born: let's build a platform that uses artificial intelligence to analyze corporate responsibility.
The spark came when the project had to be turned from an idea into a real product – and one that would work for enterprise customers. It was no easy task. The solution had to be:
feasible
scalable
technically ambitious
In practice, we harnessed artificial intelligence to understand responsibility data. We built the entire backend, user interface, and AI pipeline. We managed development according to a roadmap to ensure that the vision did not spread in all directions.
What did we learn with PlanetAI?
An AI product is not a single model – it is an entire system.
Product development requires bold decisions: which AI components will we develop ourselves, and where will we rely on ready-made solutions?
If you want to build something that doesn't yet exist, you need a team that dares to venture forth without a detailed map.
Today, several global companies use the Planet AI platform for ESG analysis.
CASE PLANET AI
"Kipinä has enabled us to take Planet AI development to a new level - qualitatively and technically. This has been crucial in enabling us to offer our services to the world's leading companies and take sustainability analysis to a whole new level."
Jussi Korpikoski
CEO, Planet Company
Check out the Planet AI artificial intelligence solution case study
What should be taken into account when productizing an AI solution?
It all starts with a problem.
If there is no real problem, there is no AI product either. With Rida and Planet, the problem was clear.
Co-creation is key.
The customer knows their problem. At Kipinä, we know the technology. The best solution is created together, not pre-designed.
AI is not a separate issue.
A good AI product doesn't feel like "artificial intelligence" – it just feels like it works.
The value of a product comes from its use.
Whether you're a doctor or a sales manager, AI is only useful if it actually saves time, improves decisions, or generates more business for salespeople.
Finally: artificial intelligence does not make products – people do.
Together with Kipinä, RIDA and Planet AI show that a successful digital product or business is not a technical coincidence.
It is a deliberate choice, a clear target group, effective implementation – and above all, people who work together.
The era of artificial intelligence has begun. But the best products are still created using the same recipe as before: by listening to customers, building, experimenting, and developing together.
Olli Laine, Kipinä Co-founder & CEO
The author is one of the visionaries of the future of Finnish digital development, a passionate advocate of data-driven leadership and one of the co-founders of Kipinä Software. In addition to his CEO hat, he wears an orienteer's headlamp and a skier's hat.