Transmetrics’ Spotlight: Bridging Differential Equations and Logistics Efficiency – Meet Milen Ivanov
11.03.2024
As part of our Transmetrics Spotlight series, we’re excited to introduce Milen Ivanov, a data scientist whose journey from a mathematics enthusiast to a key player at Transmetrics showcases the profound impact of mathematical precision in the logistics industry. Milen possesses a rich academic background, having dived into math contests since middle school before pursuing his BSc in Mathematics at the University of Warwick, UK. His academic quest continued at Brown University in the US, where he completed his Ph.D. in differential equations, focusing on pattern formation in chemical reactions.
Milen’s expertise in differential equations, a discipline closely related to the optimization tasks he now tackles, marks a seamless transition from theoretical mathematics to practical applications in data science. Joining Transmetrics eight months ago, he embarked on solving optimization problems within the transport industry, leveraging his profound understanding of mathematics to enhance operational efficiencies and significantly reduce carbon emissions through smarter logistics solutions.
Can you tell us more about your role in Transmetrics and how you decided to pursue this career?
I do optimization of flow networks in transport. Imagine the following setup: you have a list of trucks moving between cities and a list of goods in those cities, which must be delivered. What is the best way to transport the goods? What does “best” even mean? If we cannot fulfill all the orders, what is the fewest number of extra trucks we need? Will the solution scale well as we increase the number of cities? I write optimization software, which attempts to answer such questions.
The end goal is to design more efficient transportation networks, thus enabling our clients to provide dynamic and affordable services. Furthermore, I want to make transport more efficient, so that we can mitigate the carbon footprint of the logistics industry. At the end of the day, I am fortunate that my work has a positive impact on the world, which is not a given in the tech industry: sadly, the goal of many data scientists is to make people click more ads.
What drew you to work at Transmetrics, and how has your journey been so far?
I was drawn to Transmetrics by the nature of the work: optimization is like a “first cousin” to differential equations, the area of my PhD. At Transmetrics I have been fortunate to work with many exceptional people, who are very knowledgeable about their respective fields and who are happy to collaborate. Thanks to this open and friendly culture, I learned a lot about the business and I integrated myself quickly into the workflow of the company. I have many fascinating tasks to work on, and I am looking forward to making more contributions!
As a data scientist, you know that data science is a rapidly evolving field. How do you keep your skills sharp, and what areas are you currently focusing on learning more about?
Reading articles on data science, e.g. Medium, is one way to see where the field is going and gives me ideas of what libraries I should try next. The only way to keep your skills sharp is to practice and strive to deliberately improve your productive output, be it ideas, code, or documentation. Some areas where I want to improve are learning more state-of-the-art data frame libraries and time series forecasting.
I will make the almost sacrosanct omission of Large Language Models (LLM), e.g. Chat GPT, from my “to learn” list: although they are a hot topic, I am worried they will erode our democratic institutions even more than social media already has.
What are your passions or hobbies outside of work, and how do you find time for them?
I love reading of all sorts, in particular history (”The Rise and Fall of Great Powers” – Paul Kennedy), fantasy (Brandon Sanderson), and classics (Dostoevsky). I love listening to podcasts (mainly on history and neuroscience) and discussing ideas with my friends. For physical activity, I train in Shotokan karate (black belt), calisthenics, and I also like to hike in the mountains.
As to how I find time, I wonder that myself! I try to keep an organized schedule, and I do all of these because they greatly benefit one’s life! Physical activity has tremendous benefits for both physical and mental health, and reading history & classics is my way to gain a nuanced view of human nature
What advice would you give to someone looking to start a career in data science, especially in the logistics sector?
The standard data science advice is to get a good grasp of probability, statistics, and computer science, pick a project in an area you are interested in, and implement it in the language of your choice. If you are interested in the logistics sector, you should also learn about max flow/min cut theory and linear programming. Google’s OR-tools manual has an excellent collection of problems, arising in logistics, and explanations of how to apply mixed-integer programming to tackle these.
In the grander scheme of things, a data scientist wants to help people make plausible conclusions based on evidence, so understanding what the data says, and what it does not say, is key. This means that you need a good grasp of the scientific method, as well as enough domain-specific knowledge to communicate with the experts in the field you work in, be it logistics, medical image analysis, or e-commerce.