Profile: I am a machine learning scientist with 5+ years’ experience in real-time implementation of machine learning and digital signal processing algorithms. I possess updated skills, an insatiable intellectual curiosity, and the ability to mine hidden gems located within large sets of structured, semi-structured and unstructured data. Skills Programming Tools: MATLAB, C++ STL [11,14,17], Python [scikit-learn, numpy, scipy] Machine Learning: Random Forest, Naïve Bayes, Hidden Markov Models, Long Short Term Memory Networks, Support Vector machine, Regression, Convolutional and Spiking Neural Networks Signal Processing: Image Processing, Wavelet Transform, FFT, STFT, Time-Series Analysis, Cepstrum Analysis Experience Research Scientist STS Defence Limited Gosport, UK 03/2015 – Present Responsible for real-time implementation of data mining, statistical machine learning and signal processing based feature extraction algorithms for various on-going projects. Successfully led and managed research and development of UK-Innovate project, IConIC (Intelligent Condition monitoring with Integrated Communications), with a budget of £1.03 millions and 8 industrial and academic stakeholders. Developed empirical techniques for mining vibration and speed data from a ship’s engine and devised a novel methodology to detect and predict engine failures by implementing wavelet decomposition and one-class support vector machines. Devised innovative application of machine learning and statistical methods on real-time heart-beat data to predict the time to potential heart-stroke in case of medical emergency.  Research Engineer Printed Motor Works Limited Alton, UK 01/2013 – 02/2015 Provided research leadership in a team that designed and developed real-time fault-detection machine learning algorithm for next generation in-wheel motors. Provided data analytics, using advanced statistical and machine learning models, to mechanical engineering team for possible design shortcomings. Architected and implemented analytics and visualization components for device data analysis platform to predict hardware Developed current waveform extension models relying on decision trees, random forest, logistic regression and support vector machine. Education University of Portsmouth, UK PhD (Part-Time) 2016 - Continued Thesis: Combining Machine Learning and Signal Processing for Next-Gen AI algorithms University of Oxford, UK PG Cert. 2014 Major: Advance Project Management for Scientists and Engineers London Metropolitan University, UK MSc. Embedded Systems 2011 - 2012 National University of Sciences and Technology, Pakistan BEng(Hons.) Electronics Engineering 20017 - 2011
©