Welcome to the Contrast Code Challenges project page. Code challenges are an important part of our recruiting and hiring process. Every engineer/developer, designer or technologist that works at Contrast, full-time or intern submits a code challenge as part of the process to join the team.
The code challenge is designed to give candidates the ability to showcase their skills in a low stress, extended time model rather than in a conference room as part of a technical interview. Be aware that each code challenge can take longer than a day to complete, hence we give you ample time to work it. We believe that technologists work best in the comfort of their own home, coffee shop or favorite place to code or design.
Most importantly we do the challenge for you. Anyone submitting a code challenge, owns their own code, whether they put it on GitHub or another site. Our theory is that you are putting time into the interview process. We want you to get something out of the effort. What better than something to add to your portfolio?
One last important item…please do not reference Contrast on your project as we do not want to introduce the possibility of someone plagiarizing your work. We rotate projects several times a year.
We offer two options to the developer project. The first project is designed for the full stack engineer who likes to design slick, web interfaces. The second project is more suited for the backend developer who still values interactivity with users of the application, but in a different modality.
You only have to complete one project.
If you have made it this far then you’ve learned enough to know that Contrast is an application and cyber security platform. We think of security in terms of vulnerabilities, threats and attacks.
As part of this project, we would like for you to build a simple, yet elegant application that visualizes a security intelligence data feed (vulnerabilities or threat intelligence data). We ask that your UI contains at least:
It should be a clean UI written in one of the recommended JS frameworks below. You can write the service layer in the language of your choice.
We recommend that you identify one or more data feeds from one of these three sources.
Additional requirements as part of our craftsmanship initiative:
Our agent technology is the heart and soul of our product. Agent engineers need to be highly proficient in the language they are supporting. They must want to explore the internals of the language and the engine the language runs within. To really be successful you have to be willing to “hack” the language you are working. Each of the languages below provides an interface to capture a bevy of instrumentation from your web application.
This project takes can take upwards of 8-10 hours.
For this project we are asking that you build the beginnings of a Node.js agent . The agent will simply have to be configured into the runtime of a Express (or Hapi) application and perform some very basic instrumentation. We recommend that you look at existing APM technology like New Relic, AppDynamics or dynaTrace for inspiration. If you are familiar with an APM agent, then you will quickly learn how to solve a basic agent instrumentation effort.
We then want you to explore (2 out of 3) data points specifically:
Provide an interface that instruments the application to answer the questions above (preferably both to an endpoint, console and log). Ideally, the interface is a flag or switch that can be turned on/off as part of the startup of the application.
Bonus: Show metrics on the page or a separate app/webpage
For this project we are asking that you build a minimalist version of an Application Performance Monitoring (APM) library for Django (or Flask or Pyramid). The middleware should get the time that the request enters the middleware, the time that it leaves the middleware, request path, the parameter list, the MD5 hash value of the rendered output, and the current thread and process ids. This should be appended to a CSV file. The location and name of the file should be configurable by the user of the library. To make it interesting the agent, optionally, should do its MD5 calculation in compiled C or Rust code.
Goals: We are interested in how you write idiomatic and well-tested python code in a library that could be shared among many teams, particularly when some of the logic is implemented in not-python code.
virtualenv is your friend if you want to check for both 2.7 and 3.x Python compatibility.
For this project we are asking that you build a minimalist version of an Application Performance Monitoring (APM) library for Ruby on Rails. The middleware should get the time that the request enters the middleware, the time that it leaves the middleware, request path, the parameter list, the MD5 hash value of the rendered output, and the current thread and process ids. This should be appended to a CSV file. The location and name of the file should be configurable by the user of the Gem. To make it interesting, the project should add itself to the hosting Ruby on Rails application using Railties AND the agent, optionally, should do its MD5 calculation in compiled C or Rust code.
Goals: We are interested in how you write idiomatic and well-tested ruby code in a gem that could be shared among many Ruby on Rails projects particularly when some of the logic must be implemented in non-ruby code.
rbenv is your friend if you want to test in multiple ruby versions.
We have a very specific project for Java agent engineer candidates. For any applicant wanting to work on the Contrast Java agent, this project is the 2018 required project.
The project for the Java Agent Software Engineer is to create a metric-gathering extension for a web application. This “extension” should gather metrics about requests and responses served by the application and the candidate is encouraged to design this extension such that it is decoupled from the web application.
Goals: We are interested in how you write well-structured, well-tested code that needs to interact with shared mutable state. Therefore we encourage you to implement your own metric-gathering rather than use any metric-gathering capability provided by your web application framework “out of the box”. Our review’s focus will be on the metric-gathering extension and not on the web application.
We have a very specific project just for .NET agent engineers. For any applicant wanting to work on our .NET platform, this project is the 2018 required project.
Instrumenting .NET applications requires implementation of a profiler written in C++ which we consider too large for an interview project. Therefore, we ask that candidates interested in working on our .NET agent instead implement an IHttpModule.
The following project requirements are listed below:
In addition, please make sure to do the following:
At Contrast we like to play hard, work hard, and automate our Saas environment end to end. We made this project so you can showcase your skills and give us a better idea of your individual talents.
Take a look at this project here. You will need to clone the repo and send it our way when you are finished.
Bad UX is everywhere. There are hundreds of touchpoints out there that could use a little UX love. We find ourselves frustrated by door handles, interfaces, iTunes, automated checkouts, road layouts…the list goes on.
For this project, you have the opportunity to take a relatively well-known digital site or application with a less-than-stellar customer experience, create a critique of the app, and present some possible solutions with at least one polished high-fidelity mockup. Woo! You have free reign to touch on the information architecture (IA), useful additional features, workflows, terminology, interactions, visual design, etc.
We want to see your understanding and application of UX principles by demonstrating your process from concept to final solution.
Upon completion of the above, please send us your collection of artifacts and feel free to use a portfolio storage tool like Behance, Invision, etc. if you’d like. Happy designing!
Our customer facing documentation is 100% open sourced on Github. We have a strong belief that our documentation will get better with input and feedback from customers and members of the community, hence we’ve broken down barriers for the community to contribute to our content.
Our content has been built by members of our team to date. There hasn’t been a single professional author for any of our documents. We know our documentation could be vastly improved.
Candidates interested in becoming our first digital content strategist, we would like for you to do a couple of tasks designed to improve our OpenDocs, as we call it.