STATE OF THE ART
Machine Learning Conference

As a person, who spent most of his career in "soft" and administrative roles (e.g. as a project manager or an inventory manager), I asked myself multiple times: "is it possible to change careers into a more technical role? And even if yes, then at what cost and is it possible for ME?". And for a long time the answer I gave was "no, I won't be able to do it".

During this presentation, I'll share my experience of what it was like to transition (which took around 1,5 year) to the role of a Data Scientist. I'll briefly describe my journey, and how I got to where I am right now. I'll also say a few words about things that were, and still, are, challenges for me, what hindered my progress, what helped me, what I would change if I could turn back time.

Not so long ago, behind a seven mountains, behind seven cities, in the stronghold being the seat of the Lower Silesian lived an experienced data analyst. He studied the numbers with interest and eagerly talked about them. One could say that he lived a peaceful life. It would be so today, if not the day when a powerful wizard appeared before him. He gathered daredevils to travel together into the unknown for a magic orb. It was to give access to powerful might - power over all data. The young adept did not even think for a moment. He wanted to change something. He wanted adventures and knowledge.

During the presentation I will tell about my adventures related to data science, which last for a year and a half. As from a data analyst, I have transformed into a data scientist. In addition, I will share the 5 most important principles that I learned through projects that use machine learning in retail banking: credit risk prediction, income estimation, price calculation.

I'll tell you about my path from performing statistical analysis in Excel to the present day. This road ultimately took shape thanks to my involvement in subsequent projects at Veolia SSC. The common denominator of project activities and their effects is the improvement of processes. Continuous improvement is inscribed in the DNA of shared service centers, and automation is the key to the development and competitive advantage of a company. I'll put forward the postulate that in the pursuit of machine learning the path you follow is not important, but there are various obstacles to overcome depending on where you are going from. I will show that, as in the proverbial mirror, we can perceive the process approach similar to the approach using data research - the culmination of efforts is still the launch of the appropriate machine learning model.

Maybe you to would like to try your chances in the machine learning, but something is holding you back? How to start, without experience in the industry or academia? How to harness knowledge from ml/ds courses to boost your career? From over a year, I work as a data scientist, providing machine learning-based solutions, mostly for fintech and e-commerce clients, ranging from price optimization to churn prediction. However, not long ago, I also had to deal with similar difficulties. I would like to tell you about the challenges I encountered in my machine learning journey to help you deal with such problems. There is no silver bullet here - but some of the things that worked for me may also work for you.

In this talk, I want to share my story about chasing the Data Scientist/Machine Learning Engineer position. At that time, I have been struggling to find out what skill set should I acquire to be ready to take the job. I have asked myself this question for a long time. Over one year has passed since I had joined the field professionally, and my point of view about the whole process has changed. Paradoxically, I've been put in the role of a recruiter a few times. I had to think and decide what is important for the team I work with. That's why I want to share with you my understanding of Data Science. Especially to make a comparison between my past believes, and their verification by the working environment surrounding me. I hope it will help some of you to decide on the direction of your personal development.

My career in data science started, unknowingly, already in my childhood (in the 1990s), when I did my first analyses of football games and on this basis, I tried to predict results of the next games. Millions of PLN were so close! My contact with machine learning came several years later during my studies at the Wroclaw University of Technology. My master's thesis concerned the methods of data mining in medical data. Fast forward - I went to IT, where I had the opportunity to work as a tester, a software engineer, a project manager (short episode), and finally - a machine learning engineer.

During my presentation, I will tell you about the path I followed to work in data science. I will talk about what worked in my case, what I could have done better. I will also talk about my next steps to provide even move value to others. I hope my hints will speed up your career in data science.

The efficient way to meet people with whom you can do a something job, project or even business.

My career in machine learning started already at university. After graduation, I joined Estimote, a hardware startup, as a data scientist. At the moment, I don't work too much with data science topics. Does it mean that I neglect my experience? Not at all! As a data analyst and an early member of the data team in a startup from the travel industry, I currently create a space for machine learning projects. Freshly joined, I realized that in order to be able to start playing with such topics here, firstly we need to build a solid foundation - trustful data infrastructure storing the essential information. All of this requires mostly working on data mentality and assuring data quality which means lots of communication and providing support for our data engineers. Join me to learn about kicking off projects in machine learning area in the startup world, especially if you're seeking a fresh perspective.

Most of the people like to ask questions, it is the easiest way to get information, but for me not the best. In this talk,I will discuss how starting from scratch, that is without knowledge of statistics and programming, I came to my first job as a junior data scientist. Why get the first job could be hard and how I boost my chance. Why my first ML startup didn't succeed and what I learned. Finally, I also discuss what kind of feeling is it when everyone in the room believes you know that rocket-science technique which is ML, but in fact, you know where to look and how to use information.

The answer is: I don't really know, and I am not sure if there exists a one-fits-all answer. Nevertheless, I have explored some options myself and learned quite a lot en route. In my presentation, I would like to share my experience with you. I will talk about some common mistakes that I, as a young data scientist made, and how I think one can prevent them. I started my career in data science 3 years ago, fresh out of the university, not knowing much about the field. I had no idea what is important and what is not and how to make career choices. And data science is all about making choices, isn't it? ;) I believe, that my struggles from that period can be familiar to some of you, so I would like to share what I have already learned.

During the presentation, I will share my experience related to employing people for ML positions. What competences I pay attention to and what I check with the candidates. I will tell the story of two of our employees who started their careers in machine learning. Each of them has a different set of skills. I will reveal what mistakes they made and what mistakes I made in managing them.