We are looking for a senior machine learning expert and data analyst to help us design, extend, and operate financial risk scoring systems at scale. You’ll work on models and pipelines that process hundreds of terabytes of data and power decisions where accuracy, explainability, and robustness matter. This role sits at the intersection of machine learning, fintech analytics, and big-data engineering. You’ll help evolve our scoring algorithms, improve signal quality, and ensure our models remain reliable and interpretable in production environments. We’re especially interested in someone who can combine strong ML theory, hands-on data engineering, and pragmatic fintech experience.
What You’ll Do

Design and improve financial risk scoring algorithms and models.
Analyze large-scale datasets (hundreds of TBs in Elasticsearch and related systems).
Build and maintain data processing pipelines for feature generation, training, and evaluation.
Develop ML models for anomaly detection, fraud detection, credit/risk scoring, and behavioral analysis.
Validate models for accuracy, bias, stability, and drift over time.
Ensure models are explainable, auditable, and production-ready.
Work closely with engineering teams to deploy models into production systems.
Optimize performance and cost across large-scale data infrastructure.
Define metrics, dashboards, and monitoring for model performance.
Investigate edge cases and failure modes in scoring systems.

What We’re Looking For
Must-have

Senior-level experience in machine learning and data analysis. (5+ years)
Strong background in financial risk, fintech analytics, or fraud detection.
Experience building and deploying production ML models.
Strong Python ecosystem skills (NumPy, pandas, scikit-learn, PyTorch/TensorFlow, etc.).
Experience with large-scale data processing (100s of TBs).
Deep experience with Elasticsearch or similar distributed data stores.
Experience designing data pipelines (batch and/or streaming).
Strong statistical reasoning and experimentation skills.
Ability to translate business risk concepts into measurable model features.
Experience evaluating model drift, bias, and long-term stability.

Nice-to-have

Experience with real-time scoring systems.
Experience with distributed compute frameworks (Spark, Beam, Flink, etc.).
Familiarity with regulatory or compliance-sensitive environments.
Experience with graph-based risk models or transaction network analysis.
Experience building internal analytics tools or dashboards.
Knowledge of feature stores and model versioning systems.

How You Work

You think critically about data quality and signal reliability.
You design models that are robust, explainable, and production-safe.
You’re comfortable moving between analysis, modeling, and infrastructure.
You can handle messy, real-world financial data at scale.
You communicate clearly with engineers, product teams, and stakeholders.
You care about correctness and long-term maintainability.

Example Problems You Might Work On

Extending risk scoring models with new behavioral signals.
Detecting anomalous transaction patterns across massive datasets.
Improving precision/recall tradeoffs in production scoring.
Building pipelines that process and index large transaction datasets.
Designing model monitoring to detect drift and degradation.
Optimizing large-scale Elasticsearch queries and aggregations.
Combining rule-based and ML-based scoring systems.

Why Join Us

Competitive salary + performance incentives
Equity aligned with long-term growth
High ownership and direct exposure to leadership
Remote-first with global team
Health and sports benefits
Yearly international team off-sites
Work on high-impact financial risk systems.
Tackle real-world ML challenges at large scale.
Influence the architecture of our scoring and analytics platform.
Collaborate with experienced engineers and data specialists.
Own meaningful parts of our data and ML strategy.

How to Apply
Send us:

A short introduction and relevant experience.
Examples of ML or risk models you’ve worked on.
Links to projects, papers, or code (if available).
We’re particularly interested in candidates who can demonstrate experience building robust financial risk models on very large datasets and bringing them successfully into production.