Software Engineer, Intelligence / Machine Learning at Chipper Cash

Overview

• The Intelligence team is responsible for developing data-driven engineering solutions across domains such as Fraud/Risk, Identity, Product, and Growth. As a key founding member of this team, you will be expected to help shape the team’s culture, best engineering practices, and set direction/focus for tasks and execute them while being a supportive teammate.
• We are looking for engineers of generalist stock who don’t tie themselves down to specific languages or frameworks, and are motivated and curious learners that are ready to roll up their sleeves and enjoy solving problems proactively.
• Prior experience with machine learning is not needed. Although experience is preferred, we mostly care about a strong willingness to learn.

What You Will Be Doing

• You will have the rare opportunity to build on top + further enhance a modern ML/Data stack using data powered by emerging NBU (Next Billion Users) in Africa.
• The role involves taking data in its rawest form and productionizing solutions using it across the board: from exploratory analysis...to feature engineering…to building and evaluating models over this data… to integrating into Chipper’s products …to building new tooling for our core teams… and more.
• In addition to Product, you will also be collaborating with other units in our team, including: Compliance for deeper understanding of risk and fraud, Growth to help find bottlenecks in the on-boarding flow and track the growth of the app through various regional networks, Accounting to scope different financial reporting tasks, and Operations to provide a clear view into the movements of funds through the systems.

Some of the things you can expect to work on include:

• Architect end-to-end machine learning flows: imagine new feature ideas and design data pipelines ****to create new models, improve existing ones and deploy them. You will also be expected to keep up-to-date with the latest fraud-detection research. Example: performing Naive Bayes for fake name detection to use as a signal into our user risk model.
• Embed delightful and proactive experiences in our app by collaborating with Product. Example: craft suggestion chips using NLP techniques to help pre-populate payment notes for users in the Chipper app.
• Build smart tooling to empower different teams to help them make better decisions. Example: Creating a GPT-3 powered ‘analyst’ Slackbot to make data accessible throughout the team. (Yes, we have beta access!)
• Create smart reporting/alerting mechanisms to help out our operational teams and keep a tight grip on data integrity. Example: Develop an internal Retool dashboard that allows operation folks to keep track of users with inconsistent wallet balances.
• Perform exploratory data analysis in Colab/Jupyter to empower different data-driven initiatives in the company. Example: understand user behavior to validate different Product and Growth hypotheses.

Data Stack:

• PostgreSQL
• DBT
• Snowflake
• Python
• Dask
• Pinecone/Redis
• Airflow

Application Stack:

• React Native
• Node.js
• Typescript

What You Should Have

• The data stack is predominantly Python + SQL, so the ideal candidate is expected to have strong experience with both of those.
• We’re looking for talented, passionate engineers, who enjoy writing clean, well-structured, and well-tested code.
• An ideal team member will be self-driven, will enjoy owning their own projects, and will be able to constructively communicate with the rest of the team in order to brainstorm, disambiguate project specifications, and to give (and receive) feedback on everything from coding style to high level architecture and design.
• Most importantly, we’re looking for engineers who love learning.