Belvo is hiring a Senior Applied Data Scientist for its Intelligence Squad, which builds models using financial data and payments across Latin America.

Responsibilities:

  • Lead the design and selection of statistical and machine-learning models for risk scoring, fraud detection, transaction categorization, and related problems.
  • Decide which problems are worth modeling and translate business questions into modeling roadmaps and measurable business cases.
  • Use AI-assisted and agentic tools in data exploration, feature development, pipeline creation, and service implementation.
  • Build, deploy, monitor, and maintain models in production, including pipelines, APIs, drift monitoring, retraining, and retirement.
  • Design and run experiments, including A/B tests and offline evaluations.
  • Partner with Engineering, Data Research, Product, business stakeholders, and customers; communicate trade-offs and impact to technical and non-technical audiences.
  • Stay current with generative AI research and tools, and share useful findings with the team.

Requirements:

  • 5+ years of experience in data science, applied science, or a similar hands-on modeling role.
  • Experience owning models end to end in production and demonstrating measurable business impact.
  • Strong grounding in statistics and experience comparing different model types.
  • Proficiency in Python for data science, including pandas and scikit-learn, plus TensorFlow or PyTorch; and SQL.
  • Experience taking models into production, with knowledge of testing, version control, and CI/CD.
  • Experience building and evaluating production LLM applications, including tool use or agents, retrieval, and structured outputs.
  • Daily use of agentic coding tools and proficiency in written and spoken English.

Nice to have: Fintech or regulated-data experience, business or strategy background, customer-facing data science experience, MLOps, AWS model deployment, LLM evaluation tooling, and use of LLMs in data pipelines.

Tech stack: Python, pandas, scikit-learn, deep learning frameworks, SQL, cloud data warehouses, Claude, Claude Code, AWS, Datadog, CI/CD, Django, asyncio, JavaScript, Vue.js, and Sass.

Conditions and benefits: On-site in São Paulo, Brazil, or remote; flexible working hours; stock options; annual company bonus; health insurance; training budget; wellness stipends; paid time off on your birthday; remote-friendly work; team and department offsites; and additional benefits.