Luis Serrano is a Quantum AI Research Scientist at Zapata Computing. He is the author of the book Grokking Machine Learning and maintains a popular YouTube channel to explain machine learning in pedestrian terms. Luis has previously worked in machine learning at Apple and Google, and at Udacity as the head of content for AI and data science. He has a Ph.D. in mathematics from the University of Michigan, a master's and bachelor's from the University of Waterloo, and worked as a postdoctoral researcher in mathematics at the University of Quebec at Montreal.

** Show Notes**

- (2:12) Luis shared how he got excited about learning mathematics and specialized in combinatorics.
- (4:26) Luis discussed his experience studying Math for his Bachelor’s and Master’s degrees at the University of Waterloo - where he took many courses in combinatorics and engaged in undergraduate research.
- (5:59) Luis pursued his Ph.D. in Mathematics at the University of Michigan - where he worked on Schubert Calculus that intersects combinatorics and geometry (check out his Ph.D. dissertation).
- (8:45) Luis distinguished the differences between doing research in mathematics and machine learning.
- (11:33) Luis went over his time as a Postdoc Fellow and Lecturer at the University of Quebec at Montreal - where he was a member of the LaCIM lab (whose areas of research originating in Combinatorics and its relationships to Algebra and Computer Science) and taught classes in French.
- (13:47) Luis explained why he left academia and got his job as a Machine Learning Engineer at Google in 2014.
- (16:33) Luis discussed the engineering and analytical challenges he encountered as part of the video recommendations team at YouTube.
- (19:58) Luis shared lessons he learned to transition from academia to industry.
- (22:25) Luis went over his move to become the Head of Content for AI and Data Science at Udacity, alongside his online education passion.
- (26:08) Luis explained Udacity's educational approach to course content design in various nano degree programs, including Machine Learning, Deep Learning, and Data Science.
- (28:46) Luis unpacked his end-to-end process of making YouTube, where he teaches concepts in Machine Learning and Math in layman terms.
- (31:01) Luis unpacked his statement, "Humans are bad at abstraction, but great at math," from his video “You Are Much Better At Math Than You Think.”
- (34:46) Luis shared his 3 favorite Machine Learning videos: Restricted Boltzmann Machines, A Friendly Introduction to Machine Learning, and My Story with the Thue-Morse Sequence.
- (37:18) Luis discussed the data science culture at Apple, where he spent one-year teaching machine learning to the employees and doing internal consulting in AI-related projects.
- (39:06) Luis revealed his interest in quantum computing. He works as a Quantum AI Research Scientist at Zapata Computing, a quantum software company that offers computing solutions for industrial and commercial use.
- (43:19) Luis mentioned the challenges of writing “Grokking Machine Learning” - a technical book with Manning planned to be published next year - like a mystery novel.
- (46:12) Luis shared the differences between working in Silicon Valley and Canada.
- (47:50) Closing segment.

**His Contact Info**

**His Recommended Resources**

- Sebastian Thrun
- Andrew Ng
- Rana el Kaliouby
- "How Not To Be Wrong: The Power of Mathematical Thinking" by Jordan Ellenberg
- "Weapons of Math Destruction" by Cathy O'Neil

Here are the codes for free eBook copies of Luis' book **"****Grokking Machine Learning****"**: **gmldcr-D659, gmldcr-2512, gmldcr-0752, gmldcr-30A2, gmldcr-01E8. **Additionally, use the code **poddcast19 **to receive a 40% discount of all Manning products!

- In the beginning, I was actually terrible at math. It was my least favorite subject in schools. I somehow loved it, and I didn’t know, as I was always playing with puzzles and logic games, which I didn’t think were related to math.
- I realized that math is not about the formula. Once I understood the concepts, the formulas just came naturally.
- As for the sub-domain, I loved combinatorics — the study of discrete structure.
- My experience at the University of Waterloo was wonderful. They have a huge Math department with a lot of combinatorics courses. I had the opportunity to engage in the summer research program — working one-on-one with amazing professors. Ironically, I avoided probability and statistics courses then, which I use a lot now.

- I never thought about any other career choices besides academia, so doing a Ph.D. is a natural progression. The question is more about where to go for a Ph.D. The University of Michigan has a strong Math department with a lot of different topics. There was also a professor with whom I later completed my Ph.D. with.
- I worked on Schubert calculus, which is a combination of combinatorics and geometry.
- In Math, you want to prove anything and make your proof formal, regardless of its usefulness. Therefore, the work is harder than other scientific disciplines since you don’t have benchmarks to compete with. At the same time, you work on ancient problems that can date back a long time ago.

- I was in the LaCIM lab that consists of both math and computer science students and researchers. I interacted more with computer science folks and wrote papers in symmetric functions.
- I had the opportunity to teach a class in French for 100+ students. That experience forced me to learn French.

- In math academia, it’s tough to get a job as a professor. Therefore, I decided to explore other finance and consulting areas, which didn’t click for me. The last area that I thought of was tech because I didn’t think I was a good programmer.
- Then a friend referred me to work at Google. The interview was very math-oriented and full of puzzles, so I got hired there. When I looked back, I realized that I learned the right thing at the right time, but back then, I just went for whatever opportunity in front of me.

- The algorithms are not that complicated (linear models). There was so much data, making simple things to work well.
- The first challenge lies in its complexity — the computer might break, the server might go down, etc. because of the ridiculous amount of data coming in. I spent months making data coming from the logs readable.
- The second challenge is to come up with a concrete metric that can measure user satisfaction. Putting numbers into concepts and feelings was difficult.

- My dream is to democratize education and use technology to bring it to everybody. There are 4 dimensions along with this mission: (1) geographically (people who live far from schools), socio-economically (people who can’t afford education), chronologically (young and older people who want to obtain new knowledge), and personally (people with different learning styles).
- At the moment, we have a standard educational system that only fits a particular type of student. If we can build personalization systems such as Netflix and YouTube, then we can personalize education.
- Education is only for privileges in many ways. We need to bring it to everybody because they deserve it and contribute to society.

- At Udacity, we were extremely results-oriented. I would go to employers and ask the hiring managers to describe their ideal applicants. From there, we created the course projects with real data that enable students to get a job and excel in it.
- We tried to give people the narrative of machine learning and data science through our videos. We connected with the students through Slack meetings and office hours and engaged with the alumni for their feedback.

- I like to find end-to-end narratives that can be told via tales but explained as concepts. I always have 5–10 things in my head that I’m working on. I always watch videos, read articles, and explain things to family and friends until they get sick of me.
- Once there’s a click in my head, things materialize, and I can take care of the logistics. I first make the slides and animation. Then I make a pseudo-script for what I would say. Next, I record the video, which normally takes several tries. Finally, I edit it and publish it.

- As a child, I was good at math, but I didn’t know because somehow, the educational system made me think I was dumb. If I hadn’t found the Math Olympiad, I’d always thought that I am bad at math.
- What we are terrible at is abstraction. Math is everything around us: logic, reasoning, puzzle-solving, etc. But from the moment that we remove the reality from those problems and turn them into abstract concepts, they become difficult automatically.

- Quantum computing is so un-intuitive that it sparks my interest. I always want to explain it, just like how I explained Machine Learning.
- At Zapata, we have a platform for quantum algorithms that target problems that are hard for classical machine learning — such as generative modeling.
- Both the quantum algorithms and the hardware offerings are growing very fast, so I believe we’ll see practical applications of quantum computing very soon.

- Writing a book is a lot harder than making videos for me. The writing process challenged how I understand things, as I had to switch my way of thinking a lot.
- Most Machine Learning books out there are like cookbooks. For my book, I want to write it like a mystery novel.
- Given this narrative-driven approach, my editor always pushed me never to lose the readers, which was not easy. I had to find the balance between not losing the readers and keeping them guessing to make their reading experience interesting.