Software Engineer @ Permutive builds artificial intelligence that understands and recommends content at scale. We use machine learning and NLP techniques to algorithmically understand the meaning behind content and make sense of people's wide and nuanced interests based on their behaviour. From September 2015 to Present (4 months) London, United KingdomData Engineer @ Joined the demand-side platform
Software Engineer @ Permutive builds artificial intelligence that understands and recommends content at scale. We use machine learning and NLP techniques to algorithmically understand the meaning behind content and make sense of people's wide and nuanced interests based on their behaviour. From September 2015 to Present (4 months) London, United KingdomData Engineer @ Joined the demand-side platform research team - working alongside a team of data scientists to build a new real-time bidding pipeline and improve the probabilistic models involved in buying ad space on behalf of advertisers. Specific contributions involved building a real-time cookie profile store using Apache Storm for ETL and Aerospike for storage, along with working on a low-latency Java based bidding service with a requirement of handling up to 1 million requests/second.
Also built a generic A/B testing platform to be used by research and development teams across the company. This is a Scala+Akka based web service that provides an API for configuring and serving test parameters. From January 2015 to August 2015 (8 months) Software Engineer @ Worked on a redesign of the search engine component of the group's e-commerce platform. This involved building a high-availability Java service backed by Elasticsearch. The project is estimated to result in a 4% (£11m) increase in revenue for the group in its first year.
Planned and implemented the foundations of a big data pipeline for batch and real-time analytics of clickstream data, using Kafka, Elasticsearch and Hadoop. Created a front-end on top of this to visualise the impact of our search and recommendations experiments on traffic, product orders and revenue. Supervised an intern on a follow up project centered around using Apache Storm for real-time stream processing.
Built a Java based system to monitor keyword performance and perform automated keyword management for pay-per-click advertising - making extensive use of the Google AdWords API.
Led changes to the group's recommendations system - increasing the coverage and quality of on-site recommendations. These improvements are estimated to result in an additional £3.5m in revenue for the group over the next year. From July 2013 to December 2014 (1 year 6 months) Technology Industrial Placement @ Joined a delta one technology team, worked closely with the sales and trading desk to develop a search and comparison tool for exchange traded fund derivatives using Java web services and Adobe Flex. From April 2012 to September 2012 (6 months) Technology Intern @ Joined the trades completion technology team, performed renovation work on a rules engine. From June 2011 to September 2011 (4 months)
Master of Engineering (MEng), Computer Science, First Class Honours @ Imperial College London From 2009 to 2013 Josh Forman-Gornall is skilled in: Java, Distributed Systems, Python, Machine Learning, Computer Science, Algorithms, SQL, Big Data, Linux, C++, Unix, Eclipse, Spring, Maven, ElasticSearch
Looking for a different
Get an email address for anyone on LinkedIn with the ContactOut Chrome extension