Mario Grandi

Research and Analytics Lead

QA-UK Ltd (Renewable Energy Technologies Startup)•  May 2023 - June 2024•  London, United Kingdom

• Spearheaded quantitative research, planning, development, implementation, and data analysis of the full QA-UK green technologies portfolio, to find out-of-the-box solutions and directly support business decision-making. • Implemented finite element methods to CFD and ray tracing simulations for a large-scale solar/wind energy and water-collection system on AWS and HPC to reduce computational overhead and complexity by several orders of magnitude. • Effectively communicated and advised investors and partners, ensuring continued funding into all green energy research and analysis areas.

Statistical Production Analyst (Development lead)

Office for National Statistics•  September 2022 - March 2023•  Titchfield, United Kingdom

• Oversaw the development of the UK service-finance trade production and analysis team, leading the enhancements to the quantitative model software used to inform the national GDP and drive government decision-making. • Supervised refinement of Python-based pre- and post- analysis scripts, resulting in over 50% increase in productivity and setting a new standard for internal coding practices. • Detailed requirements and user stories on JIRA using AGILE methodology, orchestrating the analysis system upgrade and migration to a Python-based Hadoop environment. • Applied my experience and knowledge of Bayesian and sampling statistics to strengthen the system’s quantitative statistical analytical capabilities and operational efficacy, improving data quality and reducing overall cost. • Implemented automated analysis scripts with numpy and pandas, to streamline the production team’s work pipeline and reduce process inefficiencies saving over 2 weeks of work.

Post-Doctoral Research Fellow

University of Sussex•  August 2021 - August 2022•  Brighton, United Kingdom

• Member of a research group focused on evolving the computational capabilities of hardware and software technologies for physics experiments through R&D. • Led the R&D of a high-frequency tracking algorithm and Monte Carlo Physics simulator quantitative model, integrated via API and OOP in Python and C++ on a Linux-based distributed computing system. • Developed a statistically accurate large-scale Monte Carlo physics simulator and analysis framework in Python and C++ to test the performance of the pattern recognition algorithm. • Mentored undergraduate and postgraduate students in their final-year projects, including genetic algorithm optimization built with Scipy and pattern identification with TensorFlow.

PhD Researcher

CERN•  September 2017 - July 2021•  Brighton, United Kingdom

• Conducted quantitative research in affiliation with the ATLAS experiment at CERN focusing on processing rare signal events, of direct supersymmetric tau production in the fully hadronic decay channel, in large background noise. • Applied data science methods to carry out large-scale Bayesian statistics and inference analyses, performing linear and non-linear regressions, deep learning BDT and RNN for particle identification, Monte-Carlo simulations, feature extraction, data cleaning and transformation to maximise experimental sensitivity and contribute to the advancement of experimental physics, leading to several publications in physics journals. • Developed and implemented production-level C++, Python, and Bash software code, maintained with CI/CD, GitLab, and JIRA to perform cut-based statistical analyses and machine learning-based identification to maximise signal sensitivity over large background noise. • Underwent extensive training as a data scientist in big data analysis, machine learning, and high-performance computing techniques through the DISCnet bursary programme. • Published analyses and results in peer-reviewed physics journals and used a variety of visualisation tools to generate further insight to present at workshops, meetings, and conferences.

Data Scientist Intern

Public Health England•  June 2019 - September 2019•  Cambridge, United Kingdom

• Developed a GAN-based unsupervised deep learning AI model using scikit-learn and Tensorflow, able to generate a synthetic dataset and used to enhance patient data privacy and research capabilities. • Performed comprehensive quantitative analysis of the statistical properties of health data in Python with the Scipy and pandas libraries, to preserve statistical characteristics of the original dataset in the synthetic version, ensure data integrity, and provide actionable insights. • Coordinated project lifecycle, from concept to implementation and testing to deliver a fully functioning analysis and validation framework in SQL and Python Jupyter Notebooks, used to draw meaningful statistical conclusions and provide actionable recommendations to stakeholders and research partners. • Liaised with stakeholders, ensuring alignment with project objectives and timelines for the data generation solution. • Provided comprehensive documentation for the systems analysis, testing, and implementation for production team handover.

University of Sussex

Physics, PhD•  September 2017 - July 2021

Search for supersymmetry with the ATLAS detector at the Large Hadron Collider in final states with two hadronically decaying τ - leptons.

University of Sussex

MS•  September 2013 - August 2017•  Percentage: 70

Grade: 1st class degree with honours.