Colin McFarland
@MCFRL
Followers
1K
Following
2K
Media
18
Statuses
521
Leading teams that enable colleagues working with data and experiments through a special blend of research, tool building, and education on best practices.
Joined October 2011
It’s quite hard to explain Netflix’s unique experimentation culture. We tried our best!
netflixtechblog.com
Martin Tingley with Wenjing Zheng, Simon Ejdemyr, Stephanie Lane, Colin McFarland, Mihir Tendulkar, and Travis Brooks
3
3
24
Had a great time at @CODEConference, where I presented work that I did over the summer at Netflix with the (twitterless) Wenjing Zheng and Simon Ejdemyr on developing flexible methods to generalise or transport causal estimates from (likely non-representative) studies. 1/
Very excited about the rockstar lineup we’ve put together for @CODEConference this year! Since the conference is in two weeks, I figured I’d briefly let people know what is on the program in case you’re on the fence about attending (hint: you should attend). More info below 👇
1
5
28
👍👍 to this piece featuring @RatnaDesai1, @ch_doig and Netflix XP
2
2
18
You might enjoy our multi-part series on experimentation at Netflix. Start here 👉
netflixtechblog.com
Part 1 in a multipart series about decision making and experimentation at Netflix
0
0
9
Netflix contributions to the 2020 MIT Conference on Digital Experimentation @CODEConference. Big respect @deaneckles @sinanaral for putting this thing on and keeping the level this incredible.
netflixtechblog.com
Netflix was a proud sponsor of the 2020 Conference on Digital Experimentation (CODE). Netflix data scientists were also active…
0
2
25
We share key challenges and approaches we have been using to handle small sample size and limited pre-intervention data in quasi experiments at Netflix >
netflixtechblog.com
We discuss key challenges and approaches we’ve been using to handle small sample size and limited pre-intervention data in quasi…
0
0
5
Our Netflix Experimentation Platform (XP) team reimagined our analysis framework to enable our scientists to contribute directly to the platform. Here is their blog: https://t.co/QD03UxKID5 The XP team will be present this year at MIT CODE 2019, GHC 2019, WiML 2019. Come say hi.
2
21
51
Summary of our work so far to enable scientists to contribute directly to Experimentation Platform
Here is the epic work our XP Scientific Computation team does @netflix: https://t.co/97TKqwpvHq If you are interested do reach out, we have a few open roles: https://t.co/DNQnJCCrPO
0
2
14
Instead of hard-coding treatments, our platform enables teams to define aspects of the product in terms of feature flags or parameters, then update them on-the-fly. Interested in helping us scale and enable adaptive experiments? https://t.co/t58YXl7GJO
0
1
3
Reach me if you’re into this! “scaling new methods from working prototypes to efficient implementation on our production platforms is of primary importance, and to do so you’ll be partnering with an active mathematical engineering group”
Big chance to partner with my team to move experimentation forward at Netflix >
1
1
1
Big chance to partner with my team to move experimentation forward at Netflix >
2
4
7
"Scientists and engineers can contribute new innovations on three fronts, data, statistics, and visualizations." From Python at Netflix (features Experimentation Platform and much more) >
0
1
3
"What will be the cost of rolling out the winning cell of an AB test to all users?" Ambition levels of this team 🙌 cc / @jaswani
https://t.co/j4vWIVdcho
0
0
4
Sharing the design principles guiding the development of experimentation platform at Netflix. Interested in enabling scientists? We'd love to hear from you. https://t.co/bM8xVY3kZS
0
3
8
Improving Experimentation Efficiency with Optimal Stopping
link.medium.com
By Gang Su & Ian Yohai
0
1
13
I am absolutely here for this!
1
1
4
Our new place to share current focus, progress, open roles. Proud to be part of this incredible team, sharing more soon👌 “Accelerating and Democratizing Machine Learning & Experimentation Innovation” https://t.co/plj6W68zA4
0
2
13