R sig jobs - Jun 2012 - Data scientist // Berlin-based startup using probabilistic models in ecommerce

*Fluidshopping is a Berlin-based startup working on a customer analytics
tool for online retailers.

Customer Lifefitime Value (CLV) is the mythical 'magic number', the
amount
of money a particular customer will ever bring in. Knowing your CLV makes
it trivial to:
- optimize marketing spend for different inbound channels.
- identify your highest value customers,
- identify those in danger of never coming back.

Most shops have a 'gut feeling' of what their CLV is, but no way of
knowing
for sure. Yet this is the number that would make the largest impact to
their bottom line, if only it was possible to have it. Ah! Predicting the
future is hard, right?

It turns out CLV can be computed with surprising accuracy (10% error) at
the individual level. At Fluidshopping, we use sophisticated bayesian
models to estimate CLV. We look beyond what has happened in the past (which
you get with say google analytics). CLV changes often, due to your
marketing actions, season, or changes in the market. We retrain our models
weekly to reflect those changes, and then suggest the actions that would
make the largest impact (email campaigns, popups with offers triggered by
user behavior on your site, etc).

POST: Data scientist
===================We are looking for a data scientist. To apply, please send us
code samples.
Let us see something you build; please explain what it does and how to run
it if it?s not obvious. We prefer github projects. Do not send code samples
that do not make sense out of context. We are looking for a difficult
problem you solved, or a common problem that you solved in a novel way.
Please send a sample dataset if we?ll need it to make sense of your code.

Requirements:
------------
You need to be within 5 years of your last academic appointment, or
graduation. That is, if you graduated from university more than 5 years
ago, you are not eligible. You can also apply if you had a job in research
settings within the last 5 years. We do not care about you having a title,
but we are funded by the government, and they do. They want to see someone
with a compSci-related title. We are really sorry in advance if the
requirement excludes you, a great hacker who doesn?t have one.

Responsibilities:
-----------------
Your responsibilities will be to create, adapt, and optimize probabilistic
models. The data set is a large sample of purchasing behavior. You will
need to demonstrate experience in bayesian modeling, ideally in a setting
that resembles marketing. You will need to ?play? with parameters until we
reach state-of-the-art predictive accuracy. You should be happy navigating
the literature, learning methods on the fly, and implementing algorithms
from a paper. Of course, in a small team you will need to write
production-quality code yourself. This includes testing.

Nice to have:
-------------
- Familiarity with probability theory and modeling
- Links to any work you may have contributed to the R/open source community
(for example, your GitHub page, R packages developed, or any R project
you're currently working on)
- Links to questions asked/answered on Stack Overflow related to R
- PhD

Perks:
------
- Work from home (optional)
- Dog
- Terrace with barbecue
- In the center of the Berlin startup ecosystem
- Continuous deployment, strong testing culture

To apply, contact us at: jobs at fluidshopping.com*

-- 
Best,
-Jose

Jose Quesada, PhD. (@quesada)
CEO, http://fluidshopping.com
http://www.josequesada.name/
skype: urlwolf

	[[alternative HTML version deleted]]

*Fluidshopping is a Berlin-based startup working on a customer analytics
tool for online retailers.

Customer Lifefitime Value (CLV) is the mythical 'magic number', the
amount
of money a particular customer will ever bring in. Knowing your CLV makes
it trivial to:
- optimize marketing spend for different inbound channels.
- identify your highest value customers,
- identify those in danger of never coming back.

Most shops have a 'gut feeling' of what their CLV is, but no way of
knowing
for sure. Yet this is the number that would make the largest impact to
their bottom line, if only it was possible to have it. Ah! Predicting the
future is hard, right?

It turns out CLV can be computed with surprising accuracy (10% error) at
the individual level. At Fluidshopping, we use sophisticated bayesian
models to estimate CLV. We look beyond what has happened in the past (which
you get with say google analytics). CLV changes often, due to your
marketing actions, season, or changes in the market. We retrain our models
weekly to reflect those changes, and then suggest the actions that would
make the largest impact (email campaigns, popups with offers triggered by
user behavior on your site, etc).

POST: Data scientist
===================We are looking for a data scientist. To apply, please send us
code samples.
Let us see something you build; please explain what it does and how to run
it if it?s not obvious. We prefer github projects. Do not send code samples
that do not make sense out of context. We are looking for a difficult
problem you solved, or a common problem that you solved in a novel way.
Please send a sample dataset if we?ll need it to make sense of your code.

Requirements:
------------
You need to be within 5 years of your last academic appointment, or
graduation. That is, if you graduated from university more than 5 years
ago, you are not eligible. You can also apply if you had a job in research
settings within the last 5 years. We do not care about you having a title,
but we are funded by the government, and they do. They want to see someone
with a compSci-related title. We are really sorry in advance if the
requirement excludes you, a great hacker who doesn?t have one.

Responsibilities:
-----------------
Your responsibilities will be to create, adapt, and optimize probabilistic
models. The data set is a large sample of purchasing behavior. You will
need to demonstrate experience in bayesian modeling, ideally in a setting
that resembles marketing. You will need to ?play? with parameters until we
reach state-of-the-art predictive accuracy. You should be happy navigating
the literature, learning methods on the fly, and implementing algorithms
from a paper. Of course, in a small team you will need to write
production-quality code yourself. This includes testing.

Nice to have:
-------------
- Familiarity with probability theory and modeling
- Links to any work you may have contributed to the R/open source community
(for example, your GitHub page, R packages developed, or any R project
you're currently working on)
- Links to questions asked/answered on Stack Overflow related to R
- PhD

Perks:
------
- Work from home (optional)
- Dog
- Terrace with barbecue
- In the center of the Berlin startup ecosystem
- Continuous deployment, strong testing culture

To apply, contact us at: jobs at fluidshopping.com*

	[[alternative HTML version deleted]]