Kevin Crane Email & Phone Number

San Francisco, Ca, Us

San Francisco, Ca, Us

Payments Platform

San Francisco, Ca, Us

Infrastructure engineer, scaling PlanGrid throughout the growth of the product. Acquired by Autodesk.- Designed and led the development of a new feature flag service, improving the stability and performance of all applications that relied on the old legacy feature flag service- Developed a framework for generating customizable PDF reports from user data; designed the system architecture and integration with third-party tooling- Managed load testing and debugging to identify bottlenecks in system performance and estimate capacity requirements for the new reports feature prior to launch- Created a prototype of a caching library to allow developers to easily cache the results of expensive operations in Python without requiring additional extensive configuration

San Francisco, Ca, Us

- Backend platform engineer working on Manhattan, Twitter's real-time multi-tenant distributed database, serving millions of requests per second with low latency- Integrated a new B-Tree storage engine into Manhattan, delivering large performance improvements for read-heavy workloads, saving money and operational effort through more efficient use of hardware- Built mechanisms for cross-datacenter replication to ensure consistency and availability of data- Created automated test suites for verifying data integrity during normal read/write conditions and edge cases, as well as transition and failure-injection tools for verifying correctness during shard migrations and catastrophic node failures- Designed and led the implementation of a feature to reduce storage usage through the compaction of large binary files, deleting expired or invalid entries while keeping the full database online- Relevant Technology: Java, Python, C++, Hadoop, Pig, Scala, distributed systems, Amazon Web Services (EC2, S3, CloudFormation stacks), Jira

- Wrote a new product recommendation engine using Hadoop and Pig on an Amazon Elastic Map-Reduce cluster, configured to run daily and surface deals personalized to each customer- Designed and built a new comprehensive data logging infrastructure using open-source and cloud-based software (Apache Flume, Amazon EC2 and S3); automated deployment scripts to simplify installation and configuration- Created custom builds of Apache Flume and Avro for orchestrating this newly-logged data and supplying the product recommendation engine with up-to-date customer information- Relevant Technology: Java, Python, Hadoop/Pig, Amazon Web Services (EC2, S3, EMR), Flume

Sunnyvale, Ca, Us

- Worked on the Content and Communities Team to add several back-end features to a new product- Passed data from Ruby front-end, through Java back-end, and on to a persistent Voldemort data-store, wired together with Spring configurations- Utilized multi-faceted searches to identify unique LinkedIn members who shared a particular news article, filtering out improper results, and saving positive hits in an existing data tree- Used Pig scripts to parse through a Hadoop database, gather information about a user's industry and company, saving this data to gather news articles relevant to them- Regularly published on the LinkedIn Engineering blog (http://lnkd.in/hSSMe3 and http://lnkd.in/XzyZPP)- Participated in both regular LinkedIn Hackdays and the first Intern Hackathon

Sunnyvale, California, Us

- Built an uploading service, from design to deployment, to enable customers to post log files to Proofpoint's internal servers, enabling easy access from the support staff- Researched and began implementation of a new content management system for support documents, automated with Selenium to enable easy creation of the support site

- Developed code for genomic analysis in Python and C- Tasked with converting legacy Python to C to improve performance and compatibility

Minneapolis, Mn, Us

- Installed, configured, and tested development builds of the product, then reported bugs/errors using Bugzilla

Santa Clara, Ca, Us

Worked under the Director of Developer Relations on a special project to compare the performance of NVIDIA and Intel integrated graphics chipsets. This data was compiled into a presentation that was presented to several hundred NVIDIA directors and employees.

Master’S Degree, Software Engineering

Bs, Computer Science

Study Abroad