Senior Python Data Scientist

We’re a small and fast-growing biotech startup with an amazing set of advisors including the Broad’s Anne Carpenter, and we’re using high throughput genetic perturbation experiments in human cells, automated image segmentation and quantitation of hundreds of features for every cell, and machine learning and statistics to find new treatments for rare genetic diseases faster than anyone has previously thought possible.

There are more than 5,000 rare genetic diseases, in total affecting millions of Americans, and we aim to find treatments for 100 of them in the next 10 years.

We’re looking for an exceptional computational scientist to help lead our analysis efforts, with the following criteria:

• Several years of experience in statistics, machine learning, and software development solving problems using lots of data, preferably using python’s scientific stack; seeing the world through the lens of statistics and modeling; thorough understanding of fundamentals of machine learning such as cross-validation and learning curves, plus an ability to explore new types of data independently and get an effective guess as to what sorts of models and assumptions make sense as a starting point.
• A track record of outstanding projects, publications, or presentations that demonstrate successful application of the above talents.
• Motivation to tackle some of the most challenging data problems around, to work with other sharp and highly-motivated individuals with diverse backgrounds, and to make lots of patients’ lives dramatically better.
• Some biology background is helpful; intellectual curiosity and motivation to learn is critical.

The person in this role will work with our biologists to guide our design/experiment/analyze cycle towards getting the most impactful biological information from the most rapid and cost-effective experimental approaches. This includes researching, suggesting and testing different statistical and machine learning approaches, along with changing our experimental setup to provide more useful information with every round of experiments. We’re setting the groundwork for how we’ll design and analyze thousands of experiments in the coming years.