PyData Trójmiasto #23 x Amazon Ring

We are very happy to welcome you to 23rd edition of PyData Trójmiasto hosted by Amazon Ring Team!

When: 26th of April at 18:00.

Where: Amazon Development Centre, Olivia Prime A, Al. Grunwaldzka 472E

[Registration info - Important!]
Number of seats is limited to 60. Official registration happens here through Meetup. Access to Amazon Development Centre requires us to collect first name, last name and email address (Contact Info) to fully register you as a Guest. The list of attendees freezes 4 hours before the event starts - in case of any urgent changes please leave us a note via kontakt@pydata-trojmiasto.pl . Don't forget to bring your ID in case of the security check.

Agenda
18:00 - 18:05 - Meeting boarding
18:05 - 18:10 - A few words about PyData
18:10 - 18:50 - Recommender system: a modern approach, by Jakub Nowacki
18:50 - 19:30 - If you can't measure it - you can't improve it, by Dariusz Piotrowski
19:30 - Pizza & networking!

About talks and speakers

Talk #1
Recommender system: a modern approach
Recommender systems are by all means a quite mature topic in the AI/ML community. Nonetheless, they are of a paramount importance for business, bringing direct value for marketing, sales, customer satisfaction to name a few. In the talk I will review the modern approaches to recommender systems from both scientific and business world and point out which algorithms are useful for a particular business problem. Finally, I will present how we approach a less typical recommendation problem we have in Amazon Ring.

About speaker:
Jakub Nowacki works as Senior Data Scientist at Ring. He is maths and computers guy as well as a leader working in R&D. Graduated from Gdańsk University of Technology and University of Bristol, where he obtained PhD in Engineering Mathematics. Jakub is passionate about applying his analytical expertise to programming, particularly in the area of Machine Learning. Throughout his career, he has led R&D teams in various industries, including technology, media, pharmaceutical, and petrochemical. He's particularly interested in the application of science and mathematics to business. Jakub has a background in C/C++, but now works primarily in the JVM and Python worlds. Finally, he is an educator of data and AI/ML-related engineering and science topics.

Talk #2
If you can't measure it - you can't improve it
"If you can't measure it - you can't improve it". That's especially true in machine learning. Metrics and how we use them are shaping the direction for our models. If they are misaligned with our goals, we might end up degrading the model during development, using a suboptimal version of it or, worst of all, putting something in prod that does not belong there. This presentation focuses on how proper ways of evaluating machine learning models. I will discuss different metrics and their purposes, share a few tips on how to align the evaluation with business goals and finally discuss statistical ways to make sure we can actually trust the results we are getting.

About speaker:
Dariusz Piotrowski - Applied Scientist at Ring Amazon. I've graduated with a Master degree in Computer Science from the Gdańsk University of Technology, with a thesis about the impact of augmentation on the effectiveness of segmentation, based on the example of kidney tumor detection. During my last years at the university, I've done an internship at Amazon Cambridge working on Alexa knowledge engine and after returning to Poland I worked in a small computer vision startup as a Machine Learning Engineer. Afterwards, I came back to Amazon as an Applied Scientist and I've been working there since (3+ years). I've started in Alexa Text To Speech, where I was developing neural models optimised for embedded devices. The models I've developed will soon be powering assistants in BMW and MINI cars. Looking for new challenges I've recently moved to Amazon Ring, where I'm working on semi-automatic evaluations to accelerate our research.

About Ring
Since its founding in 2013, Ring has been on a mission to make neighborhoods safer. From the first-ever video doorbell, to the award-winning DIY Ring Alarm system, Ring's smart home security product line, as well as the Neighbors app, offer users affordable whole-home and neighborhood security. At Ring, we are committed to making home and neighborhood security accessible and effective for everyone - while working hard to bring communities together. We're driven to create high-quality reliable products that help our neighbors to protect their homes.
Ring Labs
Amazon Development Center

See you at PyData!