Hello, everyone. In this HR blog series, we are interviewing our colleagues to give you a sneak peek on what it’s like to work at HACARUS.
Today, we are talking to one of our engineers, Unseo. He was originally hired as a student intern at HACARUS – and we are thrilled to announce that he’ll be joining our team as a full-time employee from January 2021! We asked him about his studies at university, and his experience as an intern at HACARUS.
Could you tell us a little bit about yourself?
Sure. My name is Unseo Lee. I was born in Korea and raised in Japan – I’ve forgotten most of the Korean language, so my first language is Japanese.
My recent hobbies include programming, observing meerkats at the zoo, and playing the game of Go, which I’ve been addicted to for about a year. I have 1 kyu (the highest in student ranks) in the online match now – so I’m getting very close to obtaining 1 dan (master rank).
Besides that, I’ve been playing music in a band for several years and I work out often. Oh, I also like to play with Rubik’s cube, origami and so on – I like to spend time indoors.
What sparked your interest in the world of AI?
I became interested in AI and computers because of the game of Go. I used to believe that the day AI would beat humans in the game of Go would be in a distant future – mainly because the game is so complex. It was a huge shock to me that the AI system called “AlphaGo” won against the world’s top Go player in 2016. That’s when I learned about the existence of neural networks and reinforcement learning, which were inspired by the structure of the brain. This led me to wonder “how close the computer is to the brain” and became a source of my interest in the field of Artificial Intelligence.
What did you do before joining HACARUS?
First internship experience in a science lab
I participated in a research internship at the RIKEN Brain Science Institute. As part of the internship program, I performed experiments using rats. To observe the neural circuits, a viral vector (detoxified virus) is injected into the rat’s brain. After 2 weeks, the viral vector will have been spread to the nerve cells, so the fixative liquid is poured from the rat’s heart to solidify the whole body.
Since the brain is a soft organ, we need to harden it like this before taking it out. When the brain is sliced thinly and stained with a viral vector, fluorescence can be visually recognized by being exposed to infrared light. This way, we can visualize nerve connections in the brain and help research on elucidate diseases such as Alzheimer’s.
I made a simulator of bats’ movement for my graduation research. Some bats emit ultrasonic waves. By listening to the reflected sound, they can recognize what’s around them, fly in the dark, and catch moths. I experimented whether the simulator could acquire the same behavior as a bat by learning by reinforcement learning. I created a behavior simulator for bats, using the mechanism of spatial recognition by ultrasonic waves with a sound wave simulator (FDTD).
Why did you decide to become an intern at HACARUS?
I joined HACARUS as an intern around 2 years ago, in December 2018. Back in the fall of 2017, I didn’t understand the learning theory of the brain even though I was an academic research intern at a university – that’s why I started studying statistics and machine learning.
In order to study machine learning, it was natural to start writing codes by myself. This led to me becoming familiar with the programming language called Python.
I was completely self-taught at the beginning, so it was difficult for me to grasp the overview of the field related to machine learning. I decided to join HACARUS because I wanted to acquire the skills by working with the latest technology.
After working as an engineering intern at another company, I discovered HACARUS on wantedly.com in the winter of 2018. My first impression of HACARUS was that everyone is very passionate and serious about the products they are making. It was also a plus for me that the technology they are using is not completely dependent on deep learning.
How was your experience working at HACARUS as an intern?
At first, I faced some difficulties in real-life data processing. When you work on actual projects, the data is often not as neatly maintained as that of textbooks. In real life, it is rare to just apply machine learning algorithms to solve problems – you also need to have deep insight into the data you’re dealing with. There was a lot of muddy work due to trial and error of various approaches. I like to focus on one project at a time, so I felt that the working style at HACARUS was very suitable for me.
In addition, I was able to learn more about the variety of work styles. I remember that the instructions from my mentors at HACARUS were very open. In other words, I had a high degree of freedom in terms of how I achieve my goals.
At HACARUS, we think about the ways to solve the problem, investigate it, read academic papers, experiment, discuss with colleagues, and decide the next action. I think that HACARUS has a culture that allows us to “do what we truly think is right.”
Before the COVID-19 pandemic, there was a nice atmosphere that allowed us to talk frankly whenever I was in the office. Even when we were working from home, I didn’t feel any lack of communication. We actively keep in touch online – we are even trying out a “virtual office” to work more effectively with each other.
Could you tell us more about your experience on job hunting?
HACARUS was the second company I have interned for. I didn’t apply for any other internship program after HACARUS, because I truly felt like it was the best match for me.
Moreover, HACARUS specializes in “Explainable AI” and is highly conscious on this perspective. I’m someone who wants to know the reasoning behind everything – so the idea that AI should be explainable strongly resonated with me.
I did interviews with other companies though – it was a good opportunity for me to learn about other companies and opportunities. However, HACARUS was my first choice from the beginning till the end of my job-hunting experience, considering the workplace atmosphere and my field of interest. Oh, and also – it’s great working in the Tokyo R&D center.
What are your plans for the future?
I just want to keep doing an interesting job, as well as I can. To broaden my perspective and skillset, I’m currently working on mathematics and programming skills. I’m also ready to steadily acquire the skills needed to successfully deliver the results for my projects.
Thank you, Unseo for sharing your stories!
If you haven’t already, we highly recommend watching this interesting documentary about AlphaGo – the AI system that beat the top human Go player (and eventually introduced Unseo to the world of AI!).
Would you like to work with us?
HACARUS is rapidly expanding and actively hiring new talents from all over the world. If you are interested in working with a motivated and diverse team at HACARUS, please contact us here.
HACARUS INC, founded 2014 in Kyoto, Japan is the leading provider of Explainable Lightweight AI Tools, backed by Osaka Gas and Miyako Capital (Kyoto University) among others. Its solutions are used in the Medical field to enable humans to make better, faster and more reliable decisions, based on AI driven insights. Hacarus’ proprietary AI engine is built using Sparse Modeling, an AI method that understands data like a human would – by its unique key features and is far more resource, time and energy efficient when compared to Deep Learning. To learn more, visit https://hacarus.com