Staff Applied ML Engineer - Financial Crime

Salary: £66,000 - 106,000 per year

Requirements:

• We require production experience shipping deep learning models at scale, with systems serving real traffic under latency constraints.
• We require the ability to make architecture-level decisions independently, including model selection, training infrastructure, and serving strategy, and explain the reasoning and tradeoffs.
• We require experience designing ML systems with hard latency and throughput requirements, including optimization decisions such as quantization, pre-computed embeddings, and batching strategies.
• We require strong fundamentals in deep learning, including gradient dynamics, attention mechanisms, graph message-passing, and sequence modelling.
• We require a track record of influencing technical strategy across teams.
• We require Python, PyTorch or equivalent, distributed training, and ML pipeline orchestration.
• We consider experience in FinCrime, fraud detection, AML, or regulated financial services a plus.
• We consider experience with graph-based methods such as GNNs, entity resolution, and link analysis in production a plus.
• We consider foundation model fine-tuning or LLM evaluation experience a plus.
• We consider experience establishing modern ML practices in organizations scaling their ML capabilities a plus.

Responsibilities:

• We design and ship ML and deep learning models for financial crime detection, including sequence-based, graph-based, and attention-based models serving real-time decisions at scale.
• We define the architecture strategy for how we apply modern ML to risk, including model families, serving patterns, and training paradigms.
• We build reusable end-to-end pipeline patterns from experimentation through training to production deployment for future models to follow.
• We evaluate and prototype foundation model and embedding approaches for transaction representation across FinCrime domains.
• We partner with Data Science on model evaluation, experimentation design, and causal measurement where clean A/B testing is not always possible.
• We mentor engineers and data scientists on modern ML fundamentals, production best practices, and architectural decision-making.

Technologies:

• Fine-tuning
• LLM
• PyTorch
• Python
• AI
• Network

More:

We are Wise, a global technology company on a mission to make moving and managing money across borders faster, easier, and more affordable for everyone, everywhere. Our Risk ML team sits within our FinCrime organization and is building the next generation of financial crime detection using modern architectures such as deep learning, graph neural networks, and foundation models. This is a greenfield opportunity with strong investment from senior leadership, offering high autonomy, low hierarchy, and the chance to shape our direction end to end. The role is full-time, based in London with a hybrid setup, and comes with a starting salary of 145,000 to 182,000 plus RSUs and Wise benefits. We are proud of our international, diverse, equitable, and inclusive team culture.

last updated 36 week of 2026