Data Scientist

Job Description

About Currencycloud

Currencycloud is somewhere where innovative thinking and the power of technology know no limits. We know that the success of the business depends on the people who keep it moving, which is why we are looking for exceptional talent to join our international payments revolution. 

We are an open & collaborative Fintech unicorn that lives and breathes our company values:

Be Human, Better Together and Own it.

About this role...

As a data scientist, you will be working within a growing team, consisting of data engineers, analysts, and other data scientists. The role of this team is to provide a stable data management platform to help the other departments gain actionable insights from their data.

The main responsibilities include building prototype models for investigation projects, communicating any findings with stakeholders in an iterative/agile workflow, and contributing to any production workflow code base. 

Our data stack and infrastructure is built upon Snowflake and the AWS cloud. For our day-to-day workflow, we are making frequent use of SQL queries, and the scientific stack of libraries in the Python language.

What you'll get to do?

• Contributing to project development ideas - this is a particularly important aspect of the job; you will be expected to deliver creative ideas to the company, and in return, your contributions will be considered and given credit
• Collaborate with other team members in an Agile workflow: Daily stand-ups and wrap-ups, sprint planning and refinement and writing stories and epics
• Performing data integrity checks and data cleaning of raw datasets
• Carrying out exploratory data analysis (EDA) of these datasets
• Writing prototype models in an interactive development environment (Jupyter) for presentation to colleagues, and for iterative development
• Contributing to project codebases
• Version control with Git
• Writing robust and tested code
• Reviewing pull requests from colleagues

What skills and experience you'll need?

• Essential:

- Proficiency with Python, including libraries such as numpy, pandas, matplotlib/seaborn, jupyterlab/notebooks

- SQL, any flavour

- Knowledge of software design principles including version control wth Git / Github and writing Unit tests

- Building machine learning models and pipelines such as scikit-learn with Python

- hyper-parameter tuning and cross-validuation

- Interpreting the outputs and findings of the models, and communicating these results to colleagues and stakeholders

- Supervised learning and time series forecasting

- Unsupervised learning for data exploration and clustering/segmentation analysis

• Desirable, but not essential

- Experience of writing APIs: eg Flask/FastAPI

- Experience with dashboard development: eg Plotly Dash

- Experience with Big Data processing in Spark/PySpark

- Experience with other languages used in data management - Java/Scala

- Experience with deep learning frameworks: Tensorflow/PyTorch/FastAI

- Experience with deploying model pipelines into a production environment

- Experience with model lifecycle management: eg mlflow (or alternatives)