I am a Data Science professional with extensive Python (Scikit-Learn, Pandas, NumPy, SciPy, Matplotlib, Natural Language Toolkit) experience. My specialities include the design, engineering and deployment of production-worthy Machine Learning architectures for Recommender Systems. Part of this involves the use of Natural Language Processing for Feature Engineering. I also have experience with Graph Databasing tools such as
I am a Data Science professional with extensive Python (Scikit-Learn, Pandas, NumPy, SciPy, Matplotlib, Natural Language Toolkit) experience. My specialities include the design, engineering and deployment of production-worthy Machine Learning architectures for Recommender Systems. Part of this involves the use of Natural Language Processing for Feature Engineering. I also have experience with Graph Databasing tools such as Neo4J to analyse and predict User Behaviour. My favourite tools also include SQL, Java, MongoDB, R and more. I am passionate about continual and lifelong learning.
Data Scientist @ From December 2015 to Present (1 month) Data Scientist @ I headed up the Data Science efforts at Stylect where I was responsible for the processing, modelling and analysis of all the fashion and user data at Stylect. One of the main projects was the engineering and deploying of production-worthy machine learning architectures for personalised recommendations. Our main areas of concern also included: maintaining and improving the web-scraping capabilities, building in-house analytics, multivariate testing and the iterative improvements of our algorithms.
Stylect is a mobile app for women to find their perfect shoes.
TechCrunch article: http://techcrunch.com/2014/05/01/swiping-is-the-new-liking/
- Featured by Apple amongst the best new apps in 133 countries & amongst the best apps of 2014 in Italy, Germany, Austria and Poland
- Raised a total of $1M from Forward Partners and Angel Investors
- Accelerated by Oxygen Accelerator (Oct 13 - Jan 14) at the Google Campus in London.
Check out the app here: http://bit.ly/17XsodE From April 2015 to December 2015 (9 months) London, United KingdomData Science Fellow @ The ASI Fellowship enables scientists to become data scientists and engineers
- Building a recommender system for an e-commerce startup to improve their user experience and increase sales
- Developing strategies for growth hacking and user retention for a community-building startup
- Expert-led training and practical workshops in Hadoop, Spark and MapReduce as well as Statistics and Business-specific concepts From February 2015 to April 2015 (3 months) Postdoctoral Researcher @ Expanding upon an existing study of transition state searching in enzymes, primarily conducted in the TCM group in the Cavendish Laboratory, to further include an accurate description of how Van der Waals interactions are important in careful treatment of the long-ranged electrostatic effects in enzymes. The work is currently being prepared to submit for publication. From October 2014 to February 2015 (5 months) Postdoctoral Associate @ I was broadly concerned with how enzymes, which are nature's catalysts, can produce such high catalytic rate enhancement. This involved working to describe how a combination of quantum mechanics and classical physics can enhance our understanding of catalysis. My research was primarily focussing on the effects of distal residues on enzyme active-sites and the allosteric role of the protein scaffold in catalysis. I also contributed to the mentoring of graduate students and training of undergraduate research students. From July 2014 to October 2014 (4 months) Postdoctoral Research Associate @ Exploratory work concerning transition state searching within enzymes from first principles. The primary aim was to prepare the existing body of results and analysis from my PhD thesis and to submit the work for publication. The study has since been published in the Journal of Physical Chemistry Letters.
G. Lever, D. J. Cole, R. Lonsdale, K. E. Ranaghan, D. J. Wales, A. J. Mulholland, C. -K. Skylaris and M. C. Payne, J. Phys. Chem. Lett., 5, 3614, (2014)
http://pubs.acs.org/doi/abs/10.1021/jz5018703 From March 2014 to July 2014 (5 months) Cambridge, United KingdomPhD Student - Cavendish Laboratory @ My thesis was nominated as an outstanding PhD thesis by the University of Cambridge and published in Springer Theses – the “best of the best”, for which Springer bought my intellectual property.
I was a PhD student supervised by Prof. M. C. Payne FRS and Dr. D. J. Cole (EPSRC funded), working on applications of quantum mechanics to systems of biological interest including enzyme catalysis along with protein structure and dynamics.
The first part of my thesis was published in the Journal of Physics: Condensed Matter.
G. Lever, D. J. Cole, N. D. M. Hine, P. D. Haynes, and M. C. Payne. Journal of Physics: Condensed Matter, 25, 152101, (2013) (Fast Track Communication)
In addition, my collaboration with researchers in the Astrophysics department at UCL was maintained and we published further work:
M. D. J. Hollis, S. T. Balan, Greg Lever, O. Lahav, Monthly Notices of the Royal Astronomical Society, 423, 2800 (2012) From October 2010 to March 2014 (3 years 6 months) Cambridge, United KingdomR&D European PhD Seminar 2013 @ Applied technical knowledge to solve real problems found in the typical day of a Scientist or Engineer in R&D at P&G. Gained insight into the daily challenges of the business including how to make difficult and tough decisions, set priorities and communicate ideas. Took advantage of the opportunity to lead a team, practising the skills required to motivate people, via a series of exercises and case studies working with other international PhD students and trainers from P&G. From April 2013 to April 2013 (1 month) Undergraduate Researcher- London Centre for Nanotechnology @ Computational modeling of part of an enzyme: ATP synthase (Using the VASP DFT code and the linear scaling code CONQUEST), along with amino acids and proteins where experimental structures for ATP synthase crystallised around a transition state analogue are available. Funded by EPSRC.
In addition, previous computational investigations of Bismuth on the 111 surface of Silicon were completed and published in Physical Review B:
P. Studer, S. R. Schofield, Greg Lever, D. R. Bowler, C. F. Hirjibehedin and N. J. Curson. Physical Review B: Condensed Matter and Materials Physics, 84, 041306(R) (2011) From June 2010 to September 2010 (4 months) London, United KingdomUndergraduate Researcher- London Centre for Nanotechnology @ Computational investigation of Gold particles on Silicon surfaces (using the VASP DFT code) with motivation toward conducting nanoscale wires for future electronic devices smaller in size than currently technologically available. Funded by EPSRC.
In addition, continued an investigation started as an undergraduate project that I led and published the findings in conference proceedings:
S. T. Balan, G. Lever and O. Lahav. Astronomical Society of the Pacific Conference Series, 430, 122, (2010) From June 2009 to September 2009 (4 months) London, United KingdomStudent Associate @ Encouraged wider participation in higher education and raised young people’s aspirations. I took the opportunity to inspire pupils, promote maths and physics and gain a real insight into teaching as a career through classroom teaching experiences. From June 2008 to June 2008 (1 month)
Doctor of Philosophy (Ph.D.), Computational and Theoretical Biological Physics @ University of Cambridge From 2010 to 2014 MSci, Theoretical Physics (First Class Honours) @ University College London, U. of London From 2006 to 2010 Greg Lever is skilled in: Physics, Python, Machine Learning, SQL, Mathematical Modeling, Deep Learning, Data Analysis, Nanotechnology, Molecular Dynamics, LaTeX, Science, Research, Biophysics, Matlab, Density Functional Theory