London, England, United Kingdom
Working collaboratively as part of the Data Science team on a range of different projects from Revenue Forecasting to Marketing-Mix-Model of marketing data.
Offering Data Science consultancy services focusing on AI Research.
- Working collaboratively within the AI Team. - Developing predictive machine learning models. Working with complex, large-scale time-series datasets, of NYSE options and equity price/volume data, typically with the goal of forecasting price movements of those assets. - Collaborated on multiple projects and strategies across all stages of their lifecycle; from initial ideation, quantitative research, feature engineering, data wrangling, through to model selection, optimisation and performance monitoring. - Data wrangling and analysis of large-scale, time-series datasets using standard python packages (e.g. Pandas, Numpy, SKTime). Matplotlib and Seaborn for data visualization. - Machine learning models were developed using the standard statsmodels, scikit-learn and PyTorch packages, anything and everything from simple logistic-regressions and decision trees up to more complex LSTM models. - A mixture of Cython, njit and multiprocessing packages were used to make efficiency gains where possible. - The datasets and models were hosted and run in AWS cloud environments (S3 and EC2), whilst version control was handled using Git repositories. - Thorough back-testing and optimisation of these models. - Led the development of a bespoke, portfolio optimisation and visualisation script.
- Working collaboratively within the AI Team. - Rigorous assessment of the quality of our Options datasets. - Subsequent cleaning of those Options datasets.
Developed and manufactured bespoke LED-based laboratory equipment, for the catalysis of chemical reactions using various wavelengths of light; for both academic and industrial laboratories. I successfully oversaw the full product development lifecycle for two products; from initial prototyping to marketing. This involved, iterative prototype design, writing a business plan, securing grant funding (Innovate UK REACH) and pre-orders, CE certifying my products and personally manufacturing and marketing the final units.
I worked in the research group of Prof. Richard Houlston, in the Department of Genetics and Epidemiology. Using a variety of bioinformatic techniques and software to investigate the genetic risk factors of numerous cancer types. - Drove forward multiple research projects independently, as well as contributing to a nationwide multi-institutional collaborative venture, Genomics England (100,000 genomes project) - Working with large, messy, real-world datasets. - Extensive experience in data wrangling, experimental planning, data analysis and data visualisation. - These tasks were completed using R (tidyverse packages), Python (NumPy and Pandas) and Perl. I gained extensive experience in BASH scripting and Nextflow, used to produce efficient scientific pipelines. Experiments and analyses were run on a Linux based High Performance Computer systems. - Further built on my knowledge and application of statistical analysis methods. - Further built on my presentation skills, communicating results and future plans at internal meetings and updates - I co-authoured five peer reviewed journal articles. Yngvadottir et al., Human Molecular Genetics, 2022, DOI: 10.1093/hmg/ddac089, Charlie N. Saunders et al., Neuro-Oncology, 2022, DOI: 10.1093/neuonc/noab208, Charlie N. Saunders et al. Blood Advances, 2021, DOI: 10.1182/bloodadvances.2021004423, Charlie N. Saunders et al. British Journal of Cancer, 2021, DOI:10.1038/s41416-020-01083-1, Charlie N. Saunders et al. Neuro-Oncology, 2019, DOI: 10.1093/neuonc/noz209