Optimizing Search by 80% at a Software Startup

Optimizing Search by 80% at a Software Startup

Welcome to another edition of the student showcase series by Techintern.io! We've interviewed some fantastic students who have secured internships through our platform. These are opportunities to get an inside scoop on potential internships available through Techintern.io!

We’ll be walking through their experiences connecting to companies through Techintern.io, what they worked on during their internship, and any other noteworthy points from their time there.

Today, we present Zachary Tan! Zach is a student currently in his 4th year studying Mathematics with a specialization in Combinatorics and Optimization at the University of Waterloo. We’re proud to have had Zachary as an applicant through the Techintern.io platform. During our November 2020 Hiring Round, we connected him with Charm.io, a software based startup that ranks 2.8 million consumer brands across over 150 dimensions. He was offered an interview and hired as a Software Engineering Intern from January to April 2021.

How was the process of using Techintern.io?

In order to find an internship, Zach used the then nascent Techintern.io platform, so we asked how his experience was! He believed it was positive, with the process being straightforward and simplistic as students just need to fill out a form. Students answer questions about their areas and companies of interests, which Zach said was “very intuitive to use.” Companies would then reach out for interviews, which is how Zach landed his position at Charm.io. He stated he would use techintern.io again in the future.

What did you work on?

Over the course of his internship term, Zach worked on a multitude of projects, here are a few:

Search Autocomplete

Zach's main project over the course of his internship was an autocomplete feature for the search bar, which allows users to preview search results such as brands and products. This development was one of the company’s primary goals, as it would help users navigate the site faster. Through his work in architecture and engineering, Zach created a new search system which significantly improved the site's user experience in key metrics such as discoverability and engagement. Autocomplete was essential to this effort, as it allows previously less searchable products such as pillows and earrings to be discovered by new users. You can interact with Zach's work yourself here: https://charm.io/search?q=nike

Search Responsiveness

As a part of his search engine revamp, Zach was also able to tackle the problematic runtime of Charm.io’s original search, which could take upwards of 6 seconds in previous iterations. As the programs he worked with involved working with databases of billions of brands, products and other relations, it was an engineering challenge to streamline searching through this info. However, with the guidance of his team as well as his own abilities, he was able to reach an over 80% average decrease in search time, at an average between 0.5-1 seconds compared to the previous over 6 seconds. This optimized result time also included surfacing relevant autocomplete results, which created a better user experience overall.

Brand Categorization

The final big project Zach worked on was a brand categorization problem that was core to the Charm.io product. This problem involved attempting to classify brands into different operating categories, based on their existing product dimensions. Examples include classifying "McDonald's" or "Starbucks" as "Food and Beverage" companies, and "Nike" as "Sports Apparel."

At first, Zach was unfamiliar with this problem because it was closer to the realm of data science than software engineering, but in the end, he was thankful for the experience and opportunity to engage with it. Although Zach did not get to bring the project to the finish line during his time there, he gained a “better understanding of different tech fields” outside of the regular software engineering he was used to, and worked with his teammates to make progress on the problem using novel approaches.

How was your experience at Charm.io Overall?

In his own words, “they were very descriptive about the things they wanted.” This was great for new interns to help get them on the right track. Zach said it was a different experience than larger tech companies because at Charm.io, he could focus on solving problems from a user perspective instead of technical or code qualities, so he was allowed to focus on different areas of growth. This holistic, product minded approach to engineering helped him develop in his career, and he was grateful for the opportunity to experience it.

Any unique experiences to note?

Yes! It was a unique experience because he was working from Malaysia, so it was a completely flipped 12 hours difference. Of course, the remote work and flexibility of a small team helped, so he was able to discuss his hours and complete his work efficiently.

Conclusion

That was our student showcase on Zachary Tan, a Software Engineering Intern at Charm.io! To summarize:

  • Zach worked on a multitude of projects where he built features as a product minded engineer
  • He decreased search runtime from 6+ seconds to 0.5-1 seconds on average, while adding autocomplete
  • Zach had flexibility and control over his work within a small team, and was able to work effectively despite a 12 hour time difference

For Students

If you’re a student looking for your next internship, check out Techintern.io! Our common application process is designed to make applying as easy as possible, so you can find great internships like Zach. Sign up now at https://www.techintern.io/student/signup

For Companies

Are you a company looking to hire great students such as Zach? Come join us at techintern.io! We have hiring rounds approximately once a month, and we'd love to work with you. Create your posting now at https://www.techintern.io/#onboard or feel free to reach out to kingsong@techintern.io if you have any questions.