• Insights

Published on August 25, 2021 by Jenny Maat

The RheoCube team is made up of a diverse group of scientists, researchers, thinkers, and dreamers. Our last interview was with Carolien Vis, PhD, who shared some insights on life as a RheoCube Scientific Consultant. Today, we talk to Margriet Palm, a resident Scientific Programming expert on the Data Analysis and Visualization team (DaVis).

What made you interested in Scientific Programming as a career?

It’s a discipline that combines two things l like – science and programming. Starting out, I loved how I could use programming to get the computer to do what I wanted it to. Having been a scientist, I love using science, maths and other skills to solve puzzles not tackled before. Luckily, I get to do that in my daily work.

My team’s main goal is to ensure that when users create a simulation, they can do something with it. Raw data provides the foundation for learnings, but people need to understand what a simulation is telling them. That’s where I come in, making sure users can unlock the right information. It means working with large data sets that challenge you mathematically, while accounting for the memory capacity of a computer. It’s a real balancing act, which calls for a lot of creativity – combining methods and practicality to get results.  

That blend of new ideas and real-life application is a big part of what RheoCube is. We empower people, giving them access to visualizations and quantifications, so they can learn what the data is telling them. The Davis team builds the toolbox that can make it happen.

How does working with RheoCube connect to that interest?

At RheoCube, we explore the unknown, solve new challenges, and investigate things that people haven’t figured out yet. So, I get the chance to go beyond standard cases. 

I actually remember quite clearly how this job opening got my attention. I had worked with computer simulations for 10 years, and finally saw a company that wanted to make this work commercially. The potential to apply these exciting technologies in that way was fascinating. Simulations have been used in academia for a long time, but there is limited use in industry, especially for the type of simulations we do.

Today, working as part of the RheoCube team, I get the chance to focus on the tech side of things without losing the link to science. I combine both to explore the unknown.

The fact that the team is multi-disciplinary, coming from a range of different backgrounds also gives me a massive opportunity to learn. It’s an amazing group, with scientific modelers, theoretical physicists, experimental chemists, programmers, UX designers and others. I’m learning every day and have a real chance to develop.

Working as part of the RheoCube team, I get the chance to focus on the tech side of things without losing the link to science. I combine both to explore the unknown.

Being part of a scale-up lets me take responsibility. You can get ownership of what you create. Everyone is passionate about RheoCube, what it can do, and it’s potential. There’s also quick turnaround and fast feedback. I can work on a new feature, have it put into action with internal and external users and see right away how it’s making a difference.

What aspects of RheoCube do you find the most interesting, what are the most exciting elements of the product?

From the perspective of my own team, I appreciate how we help users to make sense of simulations. Unlike other experimentation equipment, users don’t push a button and simply get one graph. They get a combined analysis, a complete, understandable 3D picture of what’s going on in a complex fluid. It’s possible to change what you see, add analyses, or combine them, that’s very impactful.

I also like the way we can deliver 3D visualization but also more complicated features. Different levels of expertise can make sense of the product, from non-expert upwards. As someone who programs, I’d be put off by not being able to deliver on these expert pieces. You need to keep experts involved as they are the people who help to make the product better.

I’m also very excited by the multiscale approach we are building for the simulation engine. I’ve seen how extremely difficult this is to do when I was doing scientific research myself and it’s been tried a lot. If you go down in scale, there are more components and computational difficulty rises exponentially. It’s crucial to find ways to circumvent doing lots of computations. I’m really impressed to see how the modeling team is handling that. 

How does strong scientific programming help to make this a unique solution?

Strong scientific programming makes the product usable. Large data sets take up a lot of computer memory and any computation requires a lot of processor power. My job is to balance the usage of memory and processing power so that large data sets can be processed relatively fast. That means making decisions like; do I use memory, or write some computations to disk so they can be reused later? These smart choices keep data analysis and visualization usable, delivering results in a reasonable time range. 

There are also many tricks so get a visualization tool to show results. For example, we have a nice color blending feature. Every fluid agent in a simulation is made of different compounds, e.g. water and oil. By giving each element a separate color we can blend them and see how things are distributed in space. In fact, insights like this can only be achieved with simulations. In a lab experiment, two fluids may have identical colors and mixing would not induce color changes, but simulations let us see how they really combine.

Such deep insights let a scientist understand what’s going on, and why an experiment produces the results it does. Generally, in a lab, you try something and see if it works, then you try something else and see how that turns out. But with simulations you can test what works from the start, then validate simulation findings in the lab. This is groundbreaking. It opens the door to exciting R&D and new levels of product development.

Where do you see RheoCube positioned in the future?

I’ve always been interested in using simulations to make lab work more efficient. I would see RheoCube as being a part of that change, a key solution and a staple in the lab of tomorrow. It would be the go-to solution to save on costs and material use, while streamlining workflows.

In real life, we can’t cut through a droplet, but with simulations we can.

Even a scientist who loves working in the lab all day can benefit from working alongside someone who uses simulations. I foresee that role growing, people who were trained as lab workers starting to specialize in simulations.  

The lab of tomorrow will be different to the one we know today. It could have teams of people doing simulations and lab work. In fact, I suspect that people will specialize in either one as it’s hard to be good at both. Labs that can get those skill sets right however will have a powerful, efficient mix that will take their R&D to the next level.