Sr. Data Science (Time series & Forecasting models)

At R2, we believe that small and medium businesses are the productive engine of society. Small and medium businesses (SMBs) make up over 90% of companies in Latin America, yet they face a trillion-dollar credit gap. Our mission is to unlock SMBs’ potential by providing financial solutions that are tailored to their needs. We are reimagining the financial infrastructure of Latin America - where SMBs financial needs are satisfied without ever having to go to a bank.

R2 enables platforms in Latin America to embed financial services that SMBs can then leverage (starting with revenue-based financing). We are a tight knit team coming from organizations such as Google, Amazon, Nubank, Uber, Capital One, Mercado Libre, Globant, and J.P. Morgan. We are entering a new phase of growth following a strategic investment from Ant International, one of the world’s largest fintechs, focused on rapidly expanding our partner footprint, strengthening our credit and underwriting capabilities, and scaling our operations across multiple markets.

We are a data-first company. Machine Learning (ML) and Deep Learning (DL) are the core of our product and data is the lifeblood for all of our decision-making. As a Senior Data Scientist, you will
sit at the helm of R2 by analyzing large data from some of the leading technology platforms in the world, and deploying clever and scalable data-driven solutions that enable new financial opportunities to millions of small businesses across Latin America. Your solutions will drive critical business decisions in an automated and scalable way.

What you will do:

• Lead forecasting initiatives by designing and implementing advanced time series models to predict sales and behavioural trends for thousands of customers.
• Develop scalable machine learning solutions that power real-time decision-making across risk management, fraud detection, and product personalization.
• Collaborate cross-functionally with product managers, engineers, and business stakeholders to translate complex data challenges into actionable insights and measurable business outcomes.
• Drive innovation in fintech applications by experimenting with cutting-edge approaches (e.g., deep learning architectures, probabilistic forecasting, and transformer-based models).
• Ensure compliance and transparency in model development, aligning with industry regulations and ethical standards for financial data usage.
• Shape the data science strategy by identifying opportunities where predictive modeling can unlock new value streams and competitive advantages.

Requirements

What We're Looking For:

Background:

• You have at least 5 years of experience with machine and deep learning in a practical setting.
• You have a good understanding of fintech products, and risk management to interpret business data effectively.
• You have a strong foundation in probability, statistics, and econometrics.

Technical expertise:

• You have strong expertise in time series forecasting methods based on statistical analysis (ARIMA, SARIMA, SARIMAX, VAR, exponential smoothing, or state-space models), on Machine & Deep Learning (Random Forest, XGBoost, RNNs, LSTMs, or Transformers), or on Bayesian theory (BSTS, Prophet, or ensemble forecasting), among others.
• You have deep knowledge of machine learning techniques for sequential data (RNNs, LSTMs, GRUs, Transformers). 
• You have strong proficiency in ML/DL frameworks in Python (e.g. Tensorflow, PyTorch, Scikit-learn).
• You are comfortable consuming data through APIs, SFTP, or straight-up CSVs.
• You care about scalable machine and deep learning solutions governed by low time and space complexity algorithms and methods.
• You have experience with explainable AI, specially in the context of Deep Learning Forecasting time series methods.

Leadership & Business acumen:

• You have a data-oriented mindset: you care about getting to the bottom of how to make decisions based on data.
• You have stakeholder management experience, keeping everyone up-to-date with key findings and explaining in a non-technical way results, methodologies and processes for data-driven decision making.