Scaling Knowledge: A Look at Airbnb Data Science Team
VloĆŸit
- Äas pĆidĂĄn 26. 08. 2024
- Airbnb is a trusted community marketplace for people to list, discover, and book unique accommodations and experiences around the world. As a consumer company, data represents the voice of Airbnb's users at scale, and Data Scientists play the role of interpreter. Data Scientists at Airbnb face unique challenges due to the dynamics of their two-sided marketplace, the online and offline components of their platform and the vast surface area of their product.
In this talk, Martin Daniel, Data Scientist at Airbnb, will discuss Airbnb's emphasis on having an experimentation-centric culture and applying machine learning rigorously to address Airbnb's unique product challenges. He will show Airbnb's approach in scaling a team from 1 to 100 Data Scientists. This includes the strategy of embedding Data Scientists on every team, investing heavily in data infrastructure and tooling as well as providing broadly accessible data education to everyone in the company.
---
Go further with Le Wagon and learn HTML, CSS, ruby, Rails, Javascript, Github, APIs and much more đ during our 2 month full-time immersive bootcamp đ www.lewagon.co...
timecodes
-----------------
1:06 - intro
1:21 - What the video will be about
2:26 - "Few words" about the speaker
4:49 - Short AirBnB business story
6:58 - How did AirBnB grow after long unpopularity period: user surveys
10:05 - Data science is about considering all users
14:20 - Case 1. Marginal-return (where to invest to get the most of money?)
18:52 - Case 2. Experiments
23:35 - Legal hosting. Recognizing twins listings
26:46 - Building tools. Bycycle-condo metaphor
29:18 - Tools in AirBnB
33:02 - Knowledge repo
36:00 - Superset (about dashboards and nice visual, open-source tableua:) )
38:00 - Questions
Incredibly insightful for someone looking to go into data science! It's great to hear some specifics of the job from an actual data scientist at Airbnb.
very insightful ..
The accent is just delecious!) An the talk is interesting.
But would have been better if someone had given the microphones to people who ask quastions. Or at least if Martin had repeated the questions, because personally I didn't hear what questions were about
Your tools are making me cry relative to my current job. I really like the internal libraries, common data processing utilities and blog post with peer review oh damn. And this was 4 years ago wtf