Data Scientist
CurrentUpon completion of my probation period, I was promoted to be Plan.com's first Data Scientist!My primary responsibility is to build systems and processes that facilitate experimentation and meaningful analysis.I am using Python (FastAPI, Flask, Dash) and R (Plumber) to build Rest APIs for making automated commercial and underwriting decisions. The development responsibilities of this role are as challenging as they are rewarding. I am responsible for:- Building end points to interact with customer facing software solutions --- processing user submitted data and returning binding commercial decisions - Training machine learning algorithms to shape our risk-appetite and make automated decisions to both enhance the user experience of our products, and reduce the costs associated with the manual processing of deal profitability and risk- Refactoring and rebuilding existing systems in Python and R to be more flexible, performant, and robust than ever before. - Building modular programs and algorithms that will facilitate A/B testing, and continuous refinement and deployment of our new data-driven solutions. Allowing, my team to test hypothesis in real-time, and shape behaviour through dynamic incentives. - Deploying and integrating my APIs into our wider technology stack through containerisation using Dockerfiles to build images, and Docker Compose to network my services together.- Leading by example by writing informative commit messages and pull requests. - Building tools repositories to be used by other Python and R developers, and writing unit tests and documentation for this code.- Working with colleagues to build multi-system ETL pipelines.My overall aim within this role is to build state-of-the-art systems that will empower both myself, and the rest of my team, to test and implement new ideas to deliver the best possible experience for our customers, our partners, and our colleagues across the business!
