Extraction and collation of Gigabytes of external insurance market prices to create a database that can be used to create optimal prices for AllClear products.

- Advanced analytics: Python, SAS, SQL- Data Science: Neural Networks, regression modelling, classification & recommendation algorithms- Screen Scraping: Python Requests, Selenium, BeautifulSoup- Machine Learning: Scikit-learn ( XGBoost/LightGBM, RandomForest, KNN Clustering)

• Managing the acquisition and implementation of bureau data sources and integrating with internal data sources to improve Credit Risk strategic selections.• Creating and automating monthly key insights and presenting in Steering packs to assist strategic and tactical decisions at board level.• Senior managerial responsibility for the entire Residential Credit Risk debtor life-cycle, comprising 10 staff and reporting directly to Head of Credit Risk.

• Production, maintenance & insight of all neural network based models and forecasts using SQL & Python.• Management of forecasting strategy and implementation of latest machine learning Python tools and techniques.• Creation of MI suite to monitor forecast accuracy and flag potential data anomalies.

- Data consultancy providing analytical support for some of the biggest UK companies across Banking, Retail, Insurance, Healthcare, Telecommunications and Energy sectors.- SAS & Python software development with strong technical skills in scorecard development, data modelling, credit risk management & management information.

• Provision of analysis in SAS EG to ensure FCA compliance on revolving credit accounts.• Extraction and manipulation of internal data using SQL Teradata.• Assessing the impact on in-life credit scoring of changes to bureau data blocks.

• Supporting the delivery of a brand new company wide data platform as lead Data Analyst.• SQL extraction and manipulation of data of varying quality and from multiple sources to provide MI using SQL Server Management Studio.• Providing an audit of all the data that currently exists within the business along with CRUD documentation that catalogues where it’s used.• Organising and running data entity workshops across the business to capture where and when data is used.

• Providing strategy & SAS consultancy support over 5 years to introduce & implement an award winning Credit Risk framework within the business.• Heavy data manipulation of very large disparate datasets to enable the creation of risk scorecards for new acquisition, in-life, collections entry & fraud (non-payment and meter tampering).• The creation of an industry leading SAS based process to effectively name ‘Occupier’ accounts using bureau data.• Mentoring analysts and providing support to develop their skillsets across all stages of scorecard development.

• Provision of SAS based data manipulation for the client’s billing and payment systems to create account arrears analysis.• Consultancy credit risk support in client meetings.• Customer base profiling and modelling to classify current customers by quality and assess those likely to default at point of application.

• Creating monthly reporting processes in SAS, VBA & Excel to measure key business metrics for review at Director level.• Creation of cluster analysis to identify customer cohorts with high levels of fraudulent activity.• Providing forced churn & fraud analysis on applicant segments through different acquisition channels.

• Using SAS to create samples of different customer groups from Oracle databases for case investigation by underwriters.• Building monthly reports in very short time periods from all business areas.• Creating and implementing a TCF dashboard to measure the degree of conduct risk within Secured Credit Risk on a monthly basis.

• Working on a priority one project at Lloyds to ensure data entering numerous mortgage and credit card risk models were SOX compliant.• Using SQL & SAS to create samples from millions of server based accounts with which to compare independently sourced information.• Creation of reporting to assess the quality of the data items extracted on accuracy, completeness & consistency with a high degree of confidence.

• Developed self-referential machine learning systems to select the optimal customers for marketing campaigns based upon previous activity.• Creating logistic regression models to score a customer’s propensity to purchase, cancel & claim on insurance policies generating additional annual income of over £250k.• Data mining Oracle databases using SAS based SQL to prevent fraud activity creating annual estimated savings of £500k.• Building customer retention segmentation which splits the base into 4 key behavioural segments.