During my 6-month placement as a Business Intelligence Engineering (BIE) Intern, I had the opportunity to grow, develop and apply my skills in the Machine Learning, Data Analytics and Data Engineering space to real-world complex business problems.

​

Some of my most notable achievements include:

​

• Taking full ownership of a particular defect in the EU Inbound Supply Chain organisation by:

  • Running office hours.​

  • Improving and aligning detection precision from 91% to 98%.

  • Developing and launching a full root-cause attribution bridge.

  • Liaising with software and vendor/carrier management teams to minimise Amazon-attributed sub-defects.

• Achieving an all-time minimum defect rate with calculated savings of $315K over a 4-month period.​​

​

• Creating Support Vector Machine (SVM) models for a holistic analysis of all the different factors contributing to another defect, to challenge payment policies for 52.7% of vendor deliveries.

​

• Proposing and pushing code modifications to nullify up to a 76.7% data discrepancy between multiple tables in a pipeline, and optimising those queries to achieve a ~65% reduction in runtime.

​

• Defining the mechanism and pipeline for a new defect, as well as the codebase to integrate into 3 other defects' pipelines.

​

​

Other achievements:

​

• Achieving 2nd place globally in the intern ML challenge.

• (Fun fact) Participating in the successful Guinness World Record attempt for "Most viewers of a Mental Health Awareness lesson, live stream on a bespoke platform".

​

The delivery of these results were made possible through determination to thoroughly dive deep into the data surrounding business problems from both a technical and business perspective.

​

• Chosen among 30 U.K. scholars to participate in a course of studying AI, Data Science, 5G, Cloud Networking, IoT and Blockchain technologies.

 

• Developed and pitched a public app centered around Computer Vision AI and GPS to verify eco-friendly actions to promote sustainable lifestyles nationwide.

​

• The idea and presentation had the scalability potential to reduce 6.8M tonnes of CO2 from the U.K.'s carbon footprint.

​

• Finalist group for U.K., Lebanon and Jordan programme.

Completed a range of Insight Programmes in various companies to gain exposure in topics from Investment Banking and Asset Management to FinTech and the use of AI in fraud detection.

​

​

• Academic research under the supervision of Prof. Matthew Piggott from the Royal School of Mines, Imperial College London.

 

• Research topic: Investigating the use and potential of AI (including Machine Learning and Deep Learning) on the U.K.'s electricity grid network with a focus on the sustainability potential of the demand side.

​

• Research includes reviews of current and industrial AI applications; testing and developing of different models and techniques, e.g. Different Stability Assessments (Transient, Small-signal, etc.), Fault Detection, Nonintrusive Load Monitoring (NILM) and Load Forecasting, etc.

​​

• Particular point of interest: Improvements to LSTM-based forecasting models using transfer learning from non-grid data (e.g., training data for calibration of financial stochastic models) and grid data from France and Texas to achieve higher accuracies in the forecasting grid load in the U.K., as well as traditional ARMA/SARIMA approaches.

• Led and participated in various sustainability-focussed technical projects, working with multiple teams across the world.

 

• Software development projects involved formulating ML algorithms to optimise and reduce different control systems' energy usage and CO2 emissions.

 

• Final project: Developed (from scratch) the full stack of a data mining and analytics web app/tool that was able to eliminate 15 metric tonnes of CO2 emissions, with a worldwide scalability.

 

• Also conducted a "deep dive" into acquiring European customer data for the purpose of both:

  • training data for a neural network that computes optimal delivery routes with the minimum carbon emissions.

  • and to incorporate existing Sustainability-Tech teams in the U.S. to Europe from a commercial perspective.

• Recipient of the Power Academy Engineering Scholarship.

 

• Audited the business' progress to Net Zero over 5 years and quantified customer engagement after the launch of a Carbon Tracer app and a new scheme for the analysis of customers' energy demand.

​

• Contributed to the planning of the new Price Control Plans for 2023-2028, which included replacing 79% of the commercial van fleet to be non-carbon vehicles, as well as contributing to the forecast of demand and cost estimation for the 5-year period across all 6 power generation technologies/areas.

​

• Analysed the financial and carbon impact of remote working post-pandemic to design and propose a hybrid-working model that would reduce the business carbon footprint (BCF) by >13%.

​

• Analysed the available resources across 1335 substations to verify a carbon insetting solution that would be implemented into 70% of all U.K. substations.