Background & Research Interests

From September 2015-June 2019 I have been working in Consultancy (Capgemini & QuantumBlack/McKinsey) as a senior and principal data scientist respectively. I tackled a wide variety of problems (optimisation, regression, classification, clustering), across different geographies (Europe/Asia) and disciplines (pharma, energy, retail, fast consumer marketing goods). 

From May 2014-September 2015 I was Head of Data at Starcount, a data science startup focussing on social media analytics. There I managed the AWS infrastructure, Hadoop cluster, wrote PIG and HIVE scripts and developing machine learning algorithms using sci-kit learn.

From October 2010-March 2014, I was employed as a software developer at OpenEye Scientific Software, in Santa Fe, New Mexico. The primary objective of my work was to develop and maintain the Lexichem toolkit.

I completed my postdoctoral fellowship in Peter Murray-Rust's group at the end of June 2010. The research project lasted for two years. The objective of the project was to develop a semantic based web service using the Google Web Toolkit, Restlet, and Apache Derby database, which made use of several ontologies I constructed in the domain of polymer science. The work made use of technologies such as Atom, Eclipse, SPARQL, RDF/OWL and RDBMS.

Prior to my postdoc, I submitted my PhD in chemoinformatics (in June 2008) at the Unilever Centre for Molecular Informatics, Cambridge University. The title of my project was "Chemical Informatics of Banned Substances." Most of my research in my PhD revolved around the use of molecular similarity, descriptor computation and machine learning methods.

I have experience in virtual screening, pharmacophore generation, clustering, data mining, toxicology prediction (currently working on classification of mutagenic toxicity) and programme extensively. My favourite programming languages are: Python, Java and R. I also programme in C++ and Perl from time to time. I am now learning more web-based scripting languages such as: Javascript, php and JSP.

I am also very interested in web service development, mashups, The Open Biomedical Ontologies, the semantic web and general web programming.


For much more current and upto date information please view my wiki.

  Open Source

I am a big fan of open source software, such as the RDKit, PyTorch, deepchem, scikit-learn and tensorflow.

© 2008 Ed Cannon Spyware Link

Ed Cannon

Welcome to my home page. Here you will find information about: myself, my current research/work, my extra curricula activities and interests, my family and any recent topics in data science that have caught my eye.