Staff Applied ML Engineer - Financial Crime in London

Wise is a global technology company, building the best way to move and manage the world's money. Whether people and businesses are sending money to another country, spending abroad, or making and receiving international payments, Wise is on a mission to make their lives easier and save them money.

About the role:

Wise moves billions across borders every year. Behind every transaction is a decision: is this safe? Our ML systems make that call - at scale, in real time, across every market we operate in. Our Risk ML team is building the next generation of financial crime detection at Wise - investing in modern architectures like deep learning, graph neural networks, and foundation models to detect increasingly sophisticated fraud and money laundering patterns. We're looking for a Staff Applied ML Engineer to lead this evolution: defining the architecture strategy, shipping production neural models, and building the blueprint that scales across FinCrime domains. This is a greenfield opportunity - you'll be setting the direction for how Wise applies modern ML to financial crime risk, with strong investment and engagement from senior leadership.

How we work:

Risk ML sits within Wise's FinCrime organisation, owning the full ML and AI foundation for financial crime detection. We're scaling into three dedicated pillars - Feature Platform, Learning Loop and Risk Modelling. You'll sit in Risk Modelling, working alongside data scientists, platform engineers, product and domain experts. We operate with high autonomy and low hierarchy. You'll own problems end-to-end - from research and architecture decisions through to production deployment and impact measurement. We value engineers who shape direction, not just execute tickets.

What will you be working on:

• Designing and shipping ML and deep learning models for financial crime detection - sequence-based, graph-based, attention-based - serving real-time decisions at Wise's scale
• Defining the architecture strategy for how Wise applies modern ML to risk - which model families, which serving patterns, which training paradigms
• Building the reusable end-to-end pipeline pattern - from experimentation through training to production deployment - that future models follow
• Evaluating and prototyping foundation model and embedding approaches for transaction representation across FinCrime domains
• Partnering with Data Science on model evaluation, experimentation design and causal measurement in domains where clean A/B testing isn't always possible
• Mentoring engineers and data scientists on modern ML fundamentals, production best practices, and architectural decision-making

What do you need:

• Production experience shipping deep learning models at scale - systems serving real traffic under latency constraints
• Ability to make architecture-level decisions independently - model selection, training infrastructure, serving strategy - and explain the reasoning and tradeoffs
• Experience designing ML systems with hard latency and throughput requirements, including optimisation decisions (quantization, pre-computed embeddings, batching strategies)
• Strong fundamentals in deep learning: gradient dynamics, attention mechanisms, graph message-passing, sequence modelling
• Track record of influencing technical strategy across teams - you don't just build, you shape direction
• Python, PyTorch (or equivalent), distributed training, ML pipeline orchestration

Nice to Have:

• Experience in FinCrime, fraud detection, AML, or regulated financial services
• Experience with graph-based methods (GNNs, entity resolution, link analysis) in production
• Foundation model fine-tuning or LLM evaluation experience
• Experience establishing modern ML practices in organisations scaling their ML capabilities

We're proud to have a truly international team, and we celebrate our differences. Inclusive teams help us live our values and make sure every Wiser feels respected, empowered to contribute towards our mission and able to progress in their careers.