London, England, United Kingdom
I'm an outgoing and passionate data scientist with over eight years experience in end to end machine learning across multiple industries and use cases, as both an IC and as a team lead. I love to have fun while I work whether that be playing with new tools and algorithms or working with colleagues to do something special. The dream job would be combining data science and creativity or doing something that can make a real difference to people.
Responsibilities: - Implement real time machine learning and data driven solutions to improve the performance of the global platform. - Build machine learning solutions to improve processes outside of the platform (Customer Service/Finance). Highlights: - Setup the first feature store and built the custom pipeline to populate and update it (Sagemaker, Lambda, Dynamo) - Developed the first three production machine learning models & pipelines to deal with millions of requests for the platform increasing the click rate by over 40% (Lambda, Sagemaker, XGBoost) - Built the first two machine learning applications focused on customer service use cases, reducing CS workload by 23% and predicting future inbound volumes to aid planning (Lambda, Sagemaker, Tensorflow, BERT, Snowflake) - Setup Snowflake and migrated all processes from Postgres, reducing query times from hours to seconds. (Snowflake, Airbyte, Lambda, Snowpipe)
I sat within the data team with the key function of supporting the business with novel and interesting uses of data for the purpose of improved insights for our clients to help aide strategic business decisions. As part of my role I had the privilege of working on numerous projects and analyses: - Data mining, machine learning and AI algorithms to provide innovative solutions to data problems - Pitching bespoke solutions and data capabilities to potential customers - Scraping websites to collect, store and analyse new data - Development of statistical models to analyse data in the pursuit of finding answers within the data - Social network analysis to utilise the large scale data present on social networks - Text mining and natural language processing to analyse large scale wordy - Building a recommendation algorithm based on customers web activity - Upskilling colleagues with SQL and machine learning concepts For these projects I have been able to use a wide plethora of programming languages, technologies data sources to aide me in my work: - Python & R For analysis and the use of machine learning algorithms using the sklearn and keras libraries in python and various packages in R - Spark, BigQuery, SQL, Hive & Pig Use numerous types of data storage software and their accompanying languages to utilise all resources available for data processing - GCP (Google Cloud Platform) & AWS (Amazon Web Services) Using DataProc/EMR for hadoop clusters to analyse large scale datasets Creating EC2 Instances to enhance compute heavy analysis - Tableau, Python & R Display the data in clear and interpretative ways to present to clients Use matplotlib, igraph, ggplot2 and plotly to achieve original approaches to data visualisation problems - Google Analytics & Adobe Analytics (Omniture) Using clients log data to provide inventive solutions to their problems and advise on future applications
InDebted is an Australian fintech removing the stigma of debt collection by using a customer centric approach. I was the first data scientist hired with the main goal of integrating intelligence into their global platform. I then progressed to build out and lead the data science team. Responsibilities: - Recruit, coach and lead the data science function within InDebted. - Coordinate with the product and engineering teams to integrate the solutions into the platform. - Plan, layout and execute the data science roadmap for the company. Highlights: - Created and executed the recruitment strategy for the data science team building from one to six individuals. - Coached and mentored data team members, of ranging abilities, on both technical and soft skills. - Led over 17 data projects across multiple time zones, delivering critical solutions on schedule.
I work within the Security Engineering Analytics team as a Data Scientist. Our task was to research and develop cutting edge machine learning applications for security use cases. Responsibilities: - Research and develop machine learning applications focussed on cyber security use cases. - Evaluate the offerings of cyber security solutions in the marketplace. - Present to key non-technical stake holders in the business. - Lead upskilling of colleagues on new technologies and frameworks. - Lead the recruitment for the team, technical exercise, interviews correlating feedback. Achievements: - Developed the first purely ML Entity Behaviour Analytics Engine to detect malicious data based on both user level and population behaviour. The engine ingests raw log files and produces anomaly scores for each event along with custom dashboards to display the information to the end user. It achieves over 90% precision and recall on industry benchmark datasets. (Python, Scala, Spark, S3, Superset, Airflow, QRadar) - Built a tool to aid in mapping raw log fields to current schemas clearing the bottleneck for adding new devices to the network. (Python, tkinter) Technologies: Python, Scala, Spark, Git, HDFS, S3, Docker
Responsibilities: - Develop machine learning solutions to calculate the optimal offers to send out to customers. - Run experiments and analysis to evaluate the impact of different campaigns. Highlights: - Developed the first machine learning solution to offer customised offers to customers based on their previous transaction history with a bespoke pipeline replacing a rules based solution whilst reducing the runtime by 80% and improving conversion by 18%. (Python, Redshift, XGBoost, Optuna)