Challenge Overview
Challenge Overview
An imaginary telecommunications provider aims to improve their five-year forecasting. We have posted an Ideation challenge to explore possible approaches to that task. In this challenge, we seek ideas for technology stacks that are ideal for applications of this kind.Please see the URL for details on the solution Ideation challenge.
Contestants may submit multiple solutions. To include multiple solutions, package each in a zip file, bundle them in a single zip file, and submit the single zip file.
Background
Please read the background section for the Solution ideation linked above. In this challenge, we seek descriptions and examples of tech stacks that solve the problem nicely.Challenge Details
Suppose an ideal multi-variate model is created that effectively forecasts actual subscription behavior, validated on historical data. The results of this model would be displayed on a dashboard. In addition, users would be able to create what-if scenarios and change the input parameters of the model through the dashboard (or generally, the UI). The tool would be used by statisticians, financial employees, and executives.Your challenge: recommend a tech stack that provides an excellent solution to this need. Priority should be given to:
- Agility: the model will change quarterly
- Usability: users will have little time and seek explanable forecasts
- Feasibility: highly unusual or very rare solutions are permitted but will not be considered competitive, excepting highly unusual circumstances
- Maintenance: is it easy to deploy and use this solution? Will it be relatively easy for resourceful non-programmers to update the visualizations?
The following technologies are commonly supported but others are also permitted:
- Pivotal Cloud Foundry
- Python
- Qlik
- Aviso
- Power BI
- Tableau
- Java
- Open source solutions to this problem
We would like you to provide recommendations for the tech stack along with supporting arguments and pros and cons of your approach. At minimum please provide a 3-page document outlining your approach.
Submission Requirements
Recommend a tech stack that provides an excellent solution to this need. Recommendations that include a functioning POC (on a trivial problem) will receive higher consideration.Your submission must include a text, .doc, PPT or PDF document that includes the following sections and descriptions:
- Overview: describe your approach in “laymen’s terms”
- Methods: If you have included a POC, include an exact step by step guide to setting up and running your POC.
- Materials: List the libraries and technologies you are using or suggesting. Include URLs for downloading them. Include license specifics.
- Discussion: Explain your choice and in particular why it is a good solution for this problem.
- Alternatives: Briefly describe alternatives one should consider and under which conditions they would be more or less favorable.
- Assumptions and Risks: what are the main risks and drawbacks of this approach?
- Results: Did you implement your approach? How’d it perform? If you’re not providing an implementation, use this section to explain the EXPECTED results.
- Other: Discuss any other issues or attributes that don’t fit neatly above that you’d also like to include.
If you submit a POC:
- Open Source is allowed, but only BSD, MIT, Apache2, etc.
- Dockerized solutions are permitted
- We won’t spend a lot of time trying to get your solution to run; please be as clean and clear as possible
Judging Criteria
These submissions will be evaluated subjectively by the client based on the following criteria:- Completeness
- Does it cover all layers properly? High level architecture diagram will be helpful.
- Can the recommended stack handle moderately large datasets with minimal effort?
- Did everything present in a logical manner?
- Comparison of your recommended stack against other product(s)
- Any references provided for further reading
- Thoroughness and Depth
- All required items for the development are correctly captured and met?
- Performance recommendations
- Quality and Readability
- Easy to understand
- Tables are recommended
- Is the document clear on assumptions made?
- Anything unique that could different this submission from others?
Final Submission Guidelines
- Documentation