Adding Context to Queries: The Story Behind Adobe’s API and UI: An Elastic{ON} Reflection

Elastic{ON}16 just wrapped up and I am still trying to wind down from the excitement. Not only because I was invited to speak at this year’s conference, but because it’s one of the more engaging conferences with product announcements, user stories, and great networking. Meeting people like Shay Banon, the founder of the very tool that brought us all together, was surreal. Then add on top of that the Elastic Team being all under one roof for those few days -- it made for a lot of search power! Of course meeting people like Karel Minařík, the Ruby and the Rails SDK core contributor, who I’ve talked with over Github issues and pull requests, is always a plus. There is no other event where you can have access to so much Elastic knowledge and expertise.

With talks like All About Elasticsearch Algorithms and Data Structures and Quantitative Cluster Sizing, it’s hard not to geek out. Of course, the user story sessions do a great job of balancing out the reality of business use cases and the real problems we are working to solve. Stories from Cisco,, and HotelTonight all offered up a great insight into different ways Elasticsearch is used in the wild. However, the data nerd in me couldn’t wait to see how the New York Times stored over 160 years of its stories in Elasticsearch.

Speaking at Elastic{ON}16 was not only an honor but a great joy. Sharing the Adobe Typekit story was something that I’ve wanted to do for a while now. It’s a bit different than most of the talks given at this conference since our story is about a small cluster that handles small amounts of data, but at a very high throughput with complex queries. The introduction of Elasticsearch to our stack was not for log analysis, big data storage, and evaluation but rather for its robust query interface and stability under high throughput. I felt that many could benefit from hearing such a story and Elastic{ON}16 was the stage on which to do just that. With a diverse speakers list and even more diverse attendees, finding the right balance of content was key. The questions at the end of the talk not only affirmed that our story was one that others can relate to, but that it had answered some questions for people struggling through the same obstacles.

If you are interested in the Adobe Typekit story, stop on over and enjoy Adding Context to Queries: The Story Behind Adobe’s API and UI. In it, I share how we are able to use advanced scripting to make real-time, high-throughput queries to power our browse UI and API.

elasticon-16-sf-adobe.jpgTo watch the full Adobe presentation, click the image above or just follow this link

Marko Iskander is a Senior Computer Scientist at Adobe on the Typekit team. He has implemented large scale implementation of Elasticsearch at various Fortune 500 clients before joining the Typekit team. After joining, his first initiative was to switch the core search and filtering of data from SQL to Elasticsearch for faster query times, simpler implementations, and support of complex business queries. Next up was the API. Currently, Marko is working on pioneering the next-level implementation of user-specific prepared data.

  • We're hiring

    Work for a global, distributed team where finding someone like you is just a Zoom meeting away. Flexible work with impact? Development opportunities from the start?