We’re looking for a Data Scientist who sees experiments not as tests, but as conversations with reality. You’ll design, run, and analyze credit experiments that shape real-time lending decisions, helping millions of Brazilian entrepreneurs access fairer credit.
The Financial AI Team
- We’re part of CloudWalk’s Financial Services domain, powering money movement and credit decisions—including real-time credit engines, repayment orchestration, dynamic pricing, and collections.
- We build and run scoring models, underwriting systems, and pricing logic that keep credit decisions fast, fair, and explainable
- We push toward event-driven, AI-augmented decisioning where experiments directly shape credit limits, default rates, and merchant growth
- We believe in data-driven democratization of access to capital
- We put curiosity first—exploring before exploiting
- We solve puzzles that demand safety, compliance, explainability, and speed all at once
What You'll Do
- Design and execute experiments for credit models, with rigorous frameworks to measure business and merchant impact
- Build systematic experimentation infrastructure—metrics, statistical methodologies, and evaluation criteria for credit model performance
- Implement A/B testing systems with proper statistical power, randomization, and causal inference methods
- Analyze results from multiple model variations, translating them into clear credit policy recommendations
- Develop scalable best practices balancing statistical rigor with business speed
- Collaborate with engineering to deploy and monitor experimental models in real-time decision engines, with rollback safety nets
- Apply measurement science to link experiments to merchant success, default rates, and financial inclusion outcomes
- Bridge offline insights to production systems through careful validation and gradual rollout strategies
Technologies / Techniques Used
- Python for analysis, modeling, and statistical computing (core language in our stack)
- SQL for large-scale feature engineering on financial datasets
- Google Cloud Platform + BigQuery for analytics infrastructure
- Statistical modeling & experimental design for credit risk evaluation
- Machine learning frameworks for classification and risk modeling
- MLflow for deployment and monitoring in production
- Docker & Kubernetes for orchestration with engineering teams
What You'll Need
- Curiosity, initiative, and a bias toward experimenting and learning fast
- Strong experimental design expertise (A/B testing, causal inference, measurement frameworks)
- Statistical rigor: power analysis, bias detection, multiple testing corrections
- Python proficiency for analysis, modeling, and statistical computation
- Measurement science skills—designing metrics and building robust evaluation frameworks
- Experience with machine learning for classification and risk modeling
- SQL skills for feature engineering and large dataset analysis
- Strong communication skills in English & Portuguese, with ability to explain technical results to non-technical audiences
Nice to Have
- Experience with Google Cloud Platform and BigQuery
- Hands-on work in credit model experimentation and measurement in production fintech/digital lending environments
- MLOps experience—deployment, monitoring, and experimentation at scale
- Background or experience in applied statistics or measurement science in business contexts (economics, operations research, etc.)
Recruitment Process Outline
- Online Assessment – evaluating theory and logical reasoning
- Technical Case Study – working with real-world financial data & experiments
- Technical Interview – discussion & case presentation
- Cultural Interview – alignment with CloudWalk values If you are not willing to take an online quiz and work on a test case, do not apply.