PyLadies Dublin x PyLadies Paris October Meetup

whykay

whykay πŸ‘©πŸ»β€πŸ’»πŸˆπŸ³οΈβ€πŸŒˆ (she/her)

Posted on November 9, 2021

PyLadies Dublin x PyLadies Paris October Meetup

For our October meetup, we partnered up with PyLadies Paris!

yay dance

πŸ‘ A huge thanks to Maud for hosting us using Streamyard, I had a night off, and it was simul-streamed to PyLadies Paris and PyLadies Dublin Youtube Channels! If you missed it or want to re-watch it again, you have 2 places to check out, don't forget to πŸ‘ and subscribe and hit that πŸ”” for upcoming videos/streams.

✨ Huge shout out to our speakers: Qishan Yang and Maria Camila Remolina-Gutiérrez

🌈 Also shout out to our community partners

πŸ™Œ The PyLadies Paris Team: Maud Levy & Maria Telenczuk, and PyLadies Dublin Team: Lei Shao (and me, Vicky).

πŸ“’ Announcements from event can be found here.


Talk details

TALK 1: Research Survivor" by Qishan Yang (PyLadies Dublin)

Description

(15 mins) My research and career. My PhD. research was about data warehouse architecture modelling, investigation, and evaluation. I also did some research on novel concepts and state-of-the-art technologies for data management in the context of big data.
After my Ph.D., I joined a project as a postdoc researcher. This project used deep learning models to investigate gait data for health. It is the first system to provide gait analysis as a service generating spatial and temporal gait features from videos. Currently, I am working in a start-up company, which provides insights into what's happening and what's working (or not) for staff in a company. It also enables Leaders and HRs to make faster and better decisions.

Bio

Data Scientist at Wrky (Dublin City University). I have been using Python for a couple of years during my research and current work. Before Python, I was using c++ and Java to build up systems, websites and analyse data. But when I needed to analyse more data and run machine learning and deep learning models more effectively and efficiently, I started using Python as the main programming language.
https://www.linkedin.com/in/qishan-yang/

TALK 2: "Unraveling the mysteries of the universe with Python" by Maria Camila Remolina-GutiΓ©rrez (Pyladies Paris)

Description:

(25 mins) In this talk Maria Camila Remolina-GutiΓ©rrez will present a short overview of Python's applications in astronomy. I will touch on the most used tech stack and I will walk through some examples ranging from computational simulations to instrumentation.

Bio:

Maria Camila is an Astrophysicist and a Systems & Computing Engineer. She recently finished a master’s degree in Computer Science in Paris, France. Her favorite programming language (which was also her first) is Python, but she also really likes C, Java and Typescript. She is very interested in the fields of: cryptography, high performance computing, human-computer interaction, software engineering, data science, numerical simulations and quantum computing. In her spare time she likes to do pottery and sculpting.
https://www.linkedin.com/in/mariacamilarg


catbug dance

Next PyLadies Dublin (our last of 2021)

Hope to see you there!


Looking for Speakers for Jan 2022

detective catbug

We are looking for speakers for Tue Jan 18th, are you interested in sharing what you are doing in Python, or something cool you found.

How long is a talk?

We prefer 5/10 mins and max 30 mins (if possible).

Will PyLadies Dublin be virtual, hybrid or in-person?

We will have to see, but we are aiming to have in-person events again as soon as it's safe to do so in early 2022.

Do I have to be female to submit a talk?

No, we are open to folks to submit a talk/demo. As long as you have read and follow our Code of Conduct, and the talk is Python-related, we are looking forward to your talk submission.

Cool, cool, cool, cool... where do I submit my talk?

πŸ‘‰ Submit your talk here! Thanks in advance.

πŸ€” I've something else to ask...

Sure thing! Email us at dublin@pyladies.com

πŸ’– πŸ’ͺ πŸ™… 🚩

Join Our Newsletter. No Spam, Only the good stuff.

Sign up to receive the latest update from our blog.

Related