Celebrating the Accomplishments of Our 2019 Engineering Interns

Anna Jensen / September 13, 2019

Arnie Bhadury also contributed to this post.

As the fall semester begins and the flock of summer tech interns have settled back in at their universities, we wanted to take the time to acknowledge the many accomplishments of this year’s engineering intern class. Three interns joined our team across offices and brought with them experience and expertise in many areas of software engineering, including machine learning, databases, and computer networking.

In Redwood City, CA, interns Shayne Rothman and Ben Kussmaul worked alongside members of our platform engineering team. Both arrived from Swarthmore College and brought with them solid computer science fundamentals as well as strong communication skills and flexibility from their experiences pair programming on school assignments and lab projects. 

Ben and Shayne (Photo by Anna Jensen)

Shayne spent her summer improving deployment confidence by revamping our automated integration and validation testing pipelines for many crucial microservices within our engineering architecture. Delving into a variety of codebases, languages, and frameworks in order to develop her tests, she further developed her customer service skills and effectively created solutions for stakeholders across engineering and non-engineering teams alike. 

“This summer taught me a lot,” Shayne said in her final presentation. “I will walk away a better engineer because of the skills I have developed…I’ve been able to learn about what qualities are important for an engineer to have to be a good team member and how those teams work.”

Shayne also introduced the third-party database mapping software JOOQ to our platform, allowing us to write SQL queries by generating an interface for our Java microservices from a database schema. To assure the quality of her work, she implemented unit tests with EasyMock. By using EasyMock, which mimicked the datastores involved in server actions, Shayne could test APIs with greater accuracy without affecting the data of regression user accounts. 

If you ever find yourself in need of an HAProxy expert, look no further than Ben, who spent much of his summer devising a strategy for smarter load balancer health checks across services in our infrastructure. He proposed a new tiered architecture to meet this need, including a health-checking server sitting underneath a cache layer as well as a referring status server which would relay results to HAProxy from the cache. This architecture would reduce the time complexity of health checking significantly. Working with members of our core platform and site reliability teams, Ben was able to document and implement this solution, which he demonstrated was scalable, minimally stateful, and easily extensible with new features. 

“The culture and work environment have been awesome,” Ben said about his Flipboard experience. “Mentorship, both through my mentor Jason and through other people helping me out, was awesome. I also appreciated being able to dive into ‘real’ production – JIRA, HackerOne, pull requests, documentation – and work through the whole process.”

Photo by Anna Jensen

When he wasn’t delving deep into HAProxy documentation and teaching himself Go, Ben also worked on smaller bugs and tasks across the backend service, such as tweaks to our spam classification model, developing service web hooks with the Slack API, and helping upgrade some outdated libraries in legacy codebases. 

Outside the office, Shayne and Ben spent the summer catching the latest action film releases at the theater down the street, playing indie board games you’ve probably never heard of, and seeking out the best boba shop in the Bay Area (Shayne claims it’s Gong Cha!). 

Meanwhile in the north, our Krut Patel joined us in our Vancouver office as a co-op student from Simon Fraser University, where he is working on a Master’s degree in Visual Computing. He spent his summer working with the recommendations team on different aspects of feed generation. In his first few weeks, Krut began by attacking our engineering backlog and implementing long requested features from our topic curation team. He showed great prowess by learning technologies he hadn’t encountered before. Almost immediately, he felt quite at ease writing GraphQL queries, making changes and adding features to our internal chatbot on Slack. Within the first couple of weeks, he exposed our topic extraction metrics to our topic curators. This new feature allowed curators to quickly visualize how their changes affect our system over time. 

Krut in Vancouver office (Photo by Colin Caufield)

Fresh out of his machine learning classes, Krut enthusiastically worked on a more responsive personalization filter in our feeds. He read several state-of-the-art papers and implemented different embedding-based classifiers and improved on our existing model in production using PyTorch. While working on this project, he learned how to set up AWS Sagemaker instances and other common machine learning practices in industry. It was especially great to see Krut document all of his work for other employees to reference.

“I enjoyed working on both the early software development focused projects and later machine learning focused project,” Krut said about his experience working with us. “This experience helped me discover that my interests lie on the software development aspect of machine learning.”

Outside of his development work, Krut further developed his foodie palate. He sampled every mocha Gastown coffee shops had to offer (his favorite was JJ Bean!). When asked for the highlight of his Flipboard summer outside of work, he responded that he liked trying different beers during our biweekly team lunches.

We invited Krut to one of the local data science meetups in Vancouver, and it has been nice to see Krut going there frequently of his own accord, providing thoughtful insights to discussions and also teaching the larger community through blog posts like “Convolution Operation — Comprehensive Guide” and “MNIST Handwritten Digits Classification using a Convolutional Neural Network (CNN)”.

While we will dearly miss this summer’s interns, we are proud to celebrate their tremendous accomplishments and wish them the best in their future endeavours. We are also excited to welcome our fall SFU co-op Syed Ikram, who joins the recommendations team in Vancouver as he works on his Master’s degree in Big Data.

Interested in interning at Flipboard? While public applications for internships are currently closed, you can always send us an email at internships at flipboard.com to get on our radar for the future.

— Anna and Arnie are curating “FlipBooks” on Flipboard