Company Overview

Straddle is building the intelligence layer for modern payments—enabling smarter, faster, and more reliable financial decisions through data and machine learning. We operate at the intersection of fintech, data infrastructure, and real-time decisioning, where the systems we build directly impact transaction success, fraud detection, and customer experience.

We are a fast-moving, high-ownership team that values speed, clarity, and pragmatic execution. We believe in delivering impact quickly, iterating continuously, and building systems that scale as the business grows.

Position Overview

We are seeking a Senior Data Scientist to own the modeling strategy, experimentation, and applied intelligence that powers Straddle's core products.

This role sits at the intersection of product, data, and machine learning with responsibility for understanding business problems deeply, designing the right analytical or ML approach, and delivering models and insights that drive real financial outcomes. You will own the full lifecycle from problem framing and feature design through model evaluation, deployment collaboration, and ongoing performance monitoring.

This is a hands-on role for someone who combines strong statistical and ML fundamentals with sharp product intuition. The ideal candidate thinks in terms of business impact first, knows when a simple heuristic beats a complex model, and can identify the data gaps standing between a good model and a great one.

Essential Functions

• Partner with product, engineering, and business stakeholders to translate business problems into well-scoped data science projects

• Own modeling strategy end-to-end: problem framing, feature selection, algorithm design, training, evaluation, and iteration

• Design and build features from transactional, behavioral, and open banking data identifying where new data sources can meaningfully improve model performance

• Develop and validate models for risk scoring, fraud detection, payment decisioning, and customer segmentation

• Establish rigorous evaluation frameworks selecting appropriate metrics, building holdout/backtesting strategies, and measuring real-world model performance

• Collaborate with ML/data engineering to deploy models into batch and real-time production systems

• Monitor model performance post-deployment, detect drift, and drive retraining or redesign when needed

• Identify data quality issues, coverage gaps, and opportunities to bring in external data that strengthens product and model outcomes

• Build dashboards and analyses that surface actionable insights for product and business teams

• Contribute to the team's analytical culture by documenting methodology, sharing learnings, and raising the bar on rigor

Desired Experience & Skills

• 4+ years in applied data science, machine learning, or quantitative analytics roles

• Strong foundation in statistics, probability, and machine learning — with a clear understanding of why you choose specific algorithms, not just how to use them

• Proficiency in Python and SQL; experience with R is a plus

• Experience building and evaluating classification, regression, and ranking models in production contexts

• Demonstrated ability to translate ambiguous business questions into structured analytical approaches

• Experience with feature engineering from complex, messy, real-world data

• Familiarity with model deployment workflows and monitoring (e.g., MLflow, Databricks, CI/CD pipelines)

• Strong data intuition ability to spot issues in data quality, distribution shifts, and feature leakage

• Excellent communication skills can explain model behavior, trade-offs, and limitations to non-technical stakeholders

• Experience in fintech, payments, or fraud/risk is a strong plus but not required

Technical Expertise

• Machine learning fundamentals: supervised/unsupervised learning, ensemble methods, model selection, hyperparameter tuning, cross-validation

• Statistical analysis, hypothesis testing, and experimental design (A/B testing)

• Feature engineering and feature store design for transactional and behavioral data

• Model evaluation: precision/recall trade-offs, calibration, lift analysis, backtesting

• Model monitoring, drift detection, and performance degradation analysis

• Databricks ecosystem (Delta Lake, MLflow, Spark) preferred

• Experience working with financial, identity, or payment data at scale

• Familiarity with AI agents, LLMs, or applied GenAI for product use cases is a plus

Culture Fit

At Straddle, data science and engineering are guided by a shared philosophy:

• Speed over perfection — momentum creates opportunity; we deliver, iterate, and improve

• Ownership mentality — we don't stop at "our part"; we ensure outcomes

• Honest, data-driven thinking — we trust the data, even when it's inconvenient

• Curiosity and creativity — we ask "why," explore ideas, and challenge assumptions

• Pragmatic execution — we balance long-term scalability with immediate business impact

• Collaborative mindset — we think out loud, share context, and make each other better

We are building systems that directly impact real financial outcomes. That responsibility demands high standards, strong judgment, and a bias toward action.