Judaism and AI: a surreal story of an orthodox Jew becoming a data scientist
Shaul Solomon has a very nontrivial life story. He grew up in a progressive Chabad home in Los Angeles, where his spiritual upbringing was influenced by a strong integration of science and technology. Yet he went through 7 years of religious studies in a hardcore orthodox bubble. He did his bachelors in Israel, in an orthodox men-only Jerusalem College of Technology, studying computer science, cyber security and Torah. Even though the degree gave him both: spiritual development and technical skills, it also left him even more confused than before. The first time he even heard about AI was a year after he finished his degree from Sam Harris’ podcast discussing panpsychism and decided to take a Coursera class where he learned the fundamental elements of neural networks (gradient descent, back-propagation, etc.) Soon after he left religion and spent a year traveling, ending up in New York with no budget. To save some money for his future trips he began working as program and administrative director at a Chabad Center in New Jersey, where his uncle was the rabbi. Being less of a believer, he continued to think about artificial intelligence, coding and computer science but it was still unclear what to do next. In order to gain a fresh perspective, he quit his job and flew to northern India. Shaul was planning to spend two months traveling but couldn’t keep himself from studying. Online computer courses, inspiring podcasts on data science, Facebook groups about machine learning have easily replaced traditional education for him. But would he have known back then that a simple Google search would lead him to become a data scientist just in a year?
“I was basically applying for different data science online courses to see what it was like. That’s how the Yandex program came up. I needed to complete a test to get accepted for the Y-Data program. It’s a Python test, which includes data science and probability/statistics. I haven’t touched coding in a year and a half at that point, and the coding I did do was in C++. It was a legitimate, proper test, though, and I spent a month studying for it. It was a beautiful balance between work and holiday, I would spend weekdays studying in cafes and weekends in picturesque villages on the India-Pakistan border. In the end I got accepted in the full course. Y-Data was a game changer. It went from data science being just a fancy idea in my head to me explaining to the other Y-Data students how I approach different problems and feeling comfortable about the whole thing, even though I still have a lot to learn.”
Today Shaul works as a teaching assistant in Y-Data program. One of his past assignments on the NLP Semantic Similarity turned into a paper that was submitted last week. He also has a full time job as a data scientist in DockTech, a maritime-based enterprise. He and his team incorporate big data, statistical and machine learning models to create a digital twin of waterways for ports and canals. In his free time, Shaul dedicates himself to his passions, Mysticism, Philosophy, AI, and Poetry. Both as independent ventures and together, fusing them into a single symbiotic organism.
An independent project he works on is the NLP-based text generation model that will combine technology and spirituality. According to Shaul, the labels associated with science and religion do not exist, there are no longer two separated parties but rather equations on one side and interpretations and stories on the other.
“We want science to be as objective as possible and religion as subjective as possible; meaningful, contextual and personal. I believe AI can help us with both, but we don’t want to create a new religion, rather a “guru” who helps us get to where we need to be. AI is not a black box in which you give data and get solutions, but it’s more of a tool to make your own discoveries. As such, there is a growing trend of ‘spiritual yet non-religious’ individuals who reject the dogmatic elements of their previous faith, yet are still proponents of liberal values, equal rights, and a holistic approach to balanced living. They do, however, still find value in analyzing how Jesus/Moses/Buddha lived, for instance. I can very much relate to this idea, growing up reading all those religious texts. I do reject the dogma, but it’s undeniable that all the stories I’ve read have perennial value. I’d like to base my ideas on different aspects of different religions. Spirituality is not difficult in the tech world at all; it’s simply not talked about, which is upsetting, but also understandable. You can discuss it but you can’t quantify it. If you can’t optimize it, how do you improve it? Technology is there to help us process things faster, approach things on a bigger scale, but whether we talk about AI or religion, the human is always the one interpreting. We therefore need a human-centric view of both AI and spirituality. My idea is to create an AI that is able to take something from any religion or piece of art and put everything together into a new coherent story, and the responsibility is on the person reading it to interpret it however they want to in the end.”
It’s easy to imagine Shaul as both: a charismatic teacher, analyzing the deep meanings of sacred texts, and a vigorous student, writing lines of code. Besides dozens of books, YouTube lectures and podcasts he’s educating himself with, he also doesn’t stop taking classes at Yandex School of Data Science, and has recently signed up for an Advanced Time Series course. Community is a key factor for Shaul, whether it’s a book club, discussing the power of consciousness, or a group of data scientists solving a challenging problem.
“The idea of community associated with learning is essential. It’s fundamental for many reasons, one of them being that learning can be intimidating. Hitting a problem with your peers is encouraging. During the Y-Data course we, as a group of four people, were a part of two projects. Each of us had strengths and weaknesses, and only working together and helping each other did we manage to find success. The Y-Data alumni network is amazing. I was able to connect with several professors through it, and I still reach out to Dimitri, my mentor during the MAFAT Radar challenge. We really have a good friendship.
I am very grateful for this pivotal moment of my life when from a wandering backpacker I became a full-fledged data scientist. Y-Data provided me with the tools, the community, the whole infrastructure that I didn’t even realize I needed. The whole learning process of the Y-Data program not only gave me an independent understanding of each separate module, but also taught me to acknowledge the connections between them within the world of data science. Being able to see the whole picture whilst understanding the intricacies of its essential components is something I have to thank the Y-Data program for. I would love to see Y-Data succeed, not only because I support it, but also because I see the good that it is doing for other people and I would love to be of help along the way.”
From running a Chabad Center to working as a data scientist, Shaul has always tried to find work that will make an impact. He divides the world of tech into bad, good and neutral, and his personal aim is to make a positive influence for a humankind rather than building algorithms for ads and commercials.
“I was reading the “Unibomber’s manifesto” the other day, it’s a tremendous piece of critique of reality. It says that you can’t have a technological revolution and the betterment of society at the same time. He fairly argues how they clash, and it was a trip reading it whilst doing my job at a high-tech startup. It’s important to understand that technology is a tool, and humanity always has to be the focal point of it. The question is not what we can make technology do, but what we would like to gain from it.”
Shaul is highly influenced by literature and can not only quote Ted Kaczynski’s anti-technology essay but also easily make a page long list of the greatest minds of the 20th century that are worth getting familiar with. Among the influencers that have affected Shaul’s personal growth you can find modern writers and philosophers as well as successful computer scientists and researchers: Alan Watts, Slavoj Zizek, Allen Ginsberg, Lex Fridman, Stuart Russell, Sam Harris, Ben Goertzel, Joscha Bach, and many others. How much time will it take though when Shaul Solomon will join this company of big thinkers and become a science influencer himself? Here are a few tips from him on studying data science or self-education in general:
“There is nothing more important in education than having passion. Do something that you care about and be interdisciplinary. Broaden your field of view on the topic you care about, you’ll be surprised how much you can actually pick up from seemingly unrelated or vaguely related fields or topics. You won’t believe how much random stuff I use in my line of work from podcasts, conversations, etc. Don’t stray too much from your main road, however, staying focused is also important. Remember that money is not an absolute; you can gain value from anything. Go for apprenticeships, reach out to people, and formulate your own opinions. Stay active in general.
If we talk about data science, there are many important things to take into consideration. Firstly, promote first principle thinking. Take a step back from the jargon and the formulas; try to understand, from a logical point of view, what the problem is, what space it occupies and what do you want to get from it, from the most abstract terms to the most specific ones. Secondly, there is the wedge that we humans seem to need that is accountability. It’s an easy way of tricking one’s discipline system within a community. Having homework submissions and deadlines helps. Thirdly, one needs to be able to understand the whole system as well as its parts, as I have already mentioned. Attention becomes a resource one can use. One needs to keep up to date on where the trends are going, be careful with them, though, you never know where you mind end up”.
