John Cummins Email and Phone Number

Building serverless container things at AWS Fargate.

Working on the Data Plane for Fargate, the serverless container offering provided by AWS. Primarily focused on the integration with Firecracker. https://aws.amazon.com/blogs/containers/under-the-hood-fargate-data-plane/https://firecracker-microvm.github.io/

When the world talks about “Amazon-scale”, they usually mean the Amazon websites. My team builds the frameworks that power some of the highest-traffic, lowest-latency websites in the world. Every request is answered with content unique to each customer, built on-demand with each click from thousands of back-end inputs. For most pages this happens in milliseconds. But, it’s not fast enough. How can we go even faster? How do we provide deeper personalization? How do we empower other Amazon developers to innovate? How do we support new platforms beyond mobile - watches, the Dash, Echo?

Description:The IDEAL (Integrated Digital Event Archiving and Library - http://www.eventsarchive.org/) and (former) CTRnet (Crisis, Tragedy, and Recovery Network - http://www.ctrnet.net/) projects, funded by the US National Science Foundation (IIS – 0916733 and 1319578), have been collecting Web archive files from important events since 2008. The team currently has a prototype spreadsheet tool to portray their data to the public. This undergraduate research will augment the tool’s functionality to provide additional analytics such as heat maps and geospatial relations. This will aid social scientist attempts to extract and understand relationships among the data. To provide the best possible user experience, a variety existing technologies such as Apache SOLR, Apache Velocity, the Google Charts API, and Google’s Geocoding API will be employed to provide additional capabilities to users.Objective:Extend functionality of the IDEAL Spreadsheet Tool in order to better support social scientists.

As an intern I wrote the compile time validation for a new language under development. The language is used to enable parallel development and rendering of components on websites across the Amazon ecosystem, but most notably Amazon.com. The parallel development allows teams to deliver more quickly and the parallel rendering reduces hardware costs. Writing the compile time validation continues to have a measurable impact on the developer experience. The validation is in production and used by developers across Amazon.