Simon Thompson

I started working on ML in 1993 when I did a wildly ambitious undergrad project trying to use genetic programming to create strategies for solving blocksworld problems. It didn't really work much, but I learned a lot and it got me a place on a Ph.D. program a year later. I spent the next quite-a-few-years doing 1990's style machine learning with some excursions into behavioral simulation and cognitive architectures like SOAR before working out a thesis on distributing boosting (like, if we are strict about it) algorithms. This then got me a job as a staff researcher at BT Labs in the East of England.

I worked for BT for the next 20 odd years eventually managing the AI research programme (with the grand title of "Program Principal") and I was Head of Practice for Big Data and Customer Experience for a while as well. I worked on ML projects for customer service, network diagnositics and network performance analytics. I also worked on some research collaborations with the EU, EPSRC, many UK universities, the UK MOD and MIT. I was very lucky that I got the chance to work on a bunch of ML projects in the early days while everyone was learning, and that I got the chance to work on many ML projects with very large scale data resources while that was really new as well.

I left BT in 2019 to work as a consultant in the City of London & Cambridge. I work for a company called GFT Technologies, we deliver AI, ML and Data Science solutions in areas like Credit Risk, Model Management and Risk, Manufacturing & Transaction Monitoring. If you need help with ML let me know!