Zafar Gilani's homepage
Cambridge PhD & Alumnus, Author, AI expert, Speaker
I currently hold Data and AI Consultant position at Johnson Matthey.
Prior to this I was the Senior Principal Consultant for AI and Deep Learning at Enzen Global Solutions Ltd. I also held the Adjunct Assistant Professor position at the University of Hong Kong.
Durig my Ph.D., I was a part of the NetOS research group at the Computer Laboratory (now called the Department of Computer Science and Technology), University of Cambridge. I was supervised by Prof. Jon Crowcroft. During my Ph.D. research I worked on characterising, detecting and measuring the social cost of automated programs (bots) in online social networks (OSNs). For this purpose, I developed Stweeler. The Stweeler project was aimed at studying usage, behaviour and impact of automated agents (bots, automated entities, automated agents) in online social networks (OSNs). To measure the social cost of bots I studied their influence over content popularity, the affects of bot traffic generated and propagated over networked systems, and the types of bots that are part of our electronic social settings. For these purposes the Stweeler project collects and maintains massive amounts of Twitter data from the freely available Streaming API. Since Stweeler collects all of the tweets from the Streaming API, we can use this dataset for a number of other avenues of research, such as sentiment analysis, and sociopolitical analysis, among others.
During my graduate studies I specialised in Distributed Computing (EMDC/MSc) from Polytechnic University of Catalonia and Royal Institute of Technology. I interned at Telefonica Research Barcelona for my MS dissertation, where I worked on creating a novel technology to pre-cache (or pre-stage) mobile Web content on or near (last mile infrastructure) the user’s device in order to save network flows and costs. This technology used a 3D view of the consumed mobile Web content across location, time and popularity to deliver insights for relevant content pre-caching. This research won ACM CoNEXT 2013 best short paper award. You may note similarities of this work to predictive analytics of today, with the difference that in this work the network itself was pre-caching content rather than individual apps or platforms.
Prior to that I worked as a visiting scientist at SLAC, Stanford for the PingER project. I did my undergrad in Information Technology (BSIT) from National University of Science and Technology’s School of Electrical Engineering and Computer Science, and received Rector’s Gold Medal for my BS Dissertation. As part of my dissertation I worked on writing an InfiniBand device driver for MPJExpress - an MPI library written in Java.