About the role

Applied AI Research Scientist developing foundation models for Sardine’s agentic fraud and financial-crime platform. Building, evaluating, and deploying real-time models for banks and fintechs.

Responsibilities

Identify and scope opportunities, design rigorous experiments, and execute the roadmap for foundation model research and development
Own evaluation of foundation model performance, including offline benchmarks, time- and entity-aware holdouts, calibration, drift and degradation monitoring, and comparisons against classical baselines
Take models from data preparation and tokenization through pretraining, fine-tuning, distillation, quantization, and deployment behind real-time inference paths with tight latency budgets
Partner with Engineering on training infrastructure, GPU efficiency, feature and embedding stores, and production-scale serving
Work with client-facing teams and customers to translate model capabilities and limits into actionable decisions for risk teams
Partner with Legal, Compliance, and customer model risk teams on explainability, documentation, and governance for regulated bank and fintech customers
Scope and drive the next generation of fraud foundation models and industry-wide adoption

4+ years in applied machine learning, quantitative modeling, or ML engineering
At least one foundation model pre-trained or substantially adapted and put in front of real traffic
Hands-on self-supervised pre-training experience
Practical fine-tuning and adaptation experience
Production experience with model serving, versioning, monitoring, and rollback
Ability to self-manage and drive ambiguous applied research projects
Clear communication with partner teams across data science, engineering, product, marketing, and external partners
Strong Python
Strong SQL
Comfort preparing very large datasets
Background in fraud, AML, payments, credit, or adversarial machine learning is a plus
Experience building and evaluating LLM-based agents in production is a plus
Publications, released models, or open-source contributions in representation learning or sequence modeling are a plus
Experience with model risk management and documentation in a regulated financial environment is a plus

Generous compensation in cash and equity
Early exercise for all options, including pre-vested
Work from anywhere: Remote-first Culture
Flexible paid time off and Year-end break
Health insurance, dental, and vision coverage for employees and dependents - US and Canada specific
4% matching in 401k / RRSP - US and Canada specific
MacBook Pro delivered to your door
One-time stipend to set up a home office — desk, chair, screen, etc.
Monthly meal stipend
Monthly social meet-up stipend
Annual health and wellness stipend
Annual Learning stipend