BrightHive is an impact-driven data technology start-up company using data trusts to transform the legal, technical, and cultural ways social services providers share data, make decisions, and affect the behavioral outcomes of beneficiaries. Together with our strategic partners, we build and support open-source interoperable data infrastructure necessary for national, state, and city governments and social and civic organizations to share data more effectively and securely, and build interventions that directly impact the lives of individuals.
One of the most important responsibilities of this role is the application of research and statistical analysis to develop recommendations for program or policy changes. Particular focus areas for BrightHive are K-12 education and national and state workforce.
To help independently assess your qualifications and skill we have designed a self contained mini data science exercise, labeled “Adult Educational Attainment and Work.”
The exercise will involve a series of progressive steps, including data analysis, regression and explanation of results along with any policy recommendations as appropriate. The datasets are similar to what you might see in Strategic Initiatives.
The total start to finish time taken for this entire assignment, including the tasks, should not exceed four (4) hours but if you find yourself taking longer feel free to submit (using the instructions provided below) what you currently have and don’t stress since many different aspects of your application are holistically evaluated.
The sections that follow cover tool and time requirements as well as how to request and submit this take home assignment. We thank you for your time taken in starting this assignment and hope that you find it interesting.
The product and engineering team at BrightHive uses Python, along with Git and Docker. However, this assignment will only focus on Python and Git. Please make sure that you have the following set up and/or installed:
- Your own Github.com account
- A programming IDE of your choice
- Python 3.x installed
- Git 2.8.1+ installed
- Access to Python 3 libraries for descriptive and statistical analysis, including regression
- A Python virtual environment to work in (e.g.
pip) - An Adobe acrobat reader
We expect this assignment to typically take 4 hours although some may finish sooner and others may finish later. You will be asked to downloaded several text files, including one data set that is about 100 megabytes in size.
- Several qualitative, descriptive analyses
- Several quantitative results written to a file
- Instructions for how to reproduce your results
The assignment is a public GitHub repository that you will clone. Your work will be done on your own private or public fork of the public repository. The assignment consists of a series of analysis tasks across a single data set. When you wish BrightHive to review your fork send a contributor invite to [email protected]. This makes a timestamp of when you are finished. Be prepared to discuss your approach in a possible follow up call.
We will send you an email containing a link to the github repository. You are free to pick any day or time that works for you but please try to start and complete the 4 hour assignment within a week of receiving the link. If something unexpected comes up and you are unable to start the data challenge please email [email protected].
The assignment is detailed in ./assignment/README.md and may be viewed in your web browswer. It contains 6 tasks related to a fictional company, called ACME Education, in collaboration with Strategic Initiatives.
In your repository add [email protected] as a contributing member with at least read access. This lets us know that you are finished with the assignment. Please make your repository private so that others completing the same assignment can not peek at your hard earned analysis. Please follow up with a quick email to [email protected] and indicate that you are finished with assignment.
BrightHive Strategic Initiatives, Product and Engineering will review your submission and contact you if we wish to continue with next steps.