AI Engineer 4 (MLX, Agentic AI, Gen AI workflows, GraphRAG)

Description

The company is building AI systems meant to be responsible and reliable, and has long used machine learning for real-time, personalised customer experiences. Examples include alerting customers to unusual charges and answering their questions in real time.

The Intelligent Foundations and Experiences (IFX) team sets the company's AI direction. It works with partners across the company on AI science and engineering, and builds and ships proprietary solutions that sit at the core of the business and reach millions of customers. Its models and platforms let other teams add AI to their products in responsible, scalable ways.

Responsibilities

• Work with a cross-functional group of engineers, research scientists, technical program managers and product managers to ship AI-powered products that change how associates work and how customers interact with the company.
• Design, build, test, deploy and support AI software, including foundation model training, LLM inference, agents and multi-agent workflows, similarity search, guardrails, model evaluation, experimentation, governance and observability.
• Use a wide range of open source and SaaS AI tools, such as AWS Ultraclusters, Hugging Face, vector databases and PyTorch.
• Research and introduce advanced foundation model optimisation methods that improve scalability, cost, latency and throughput of large production AI systems.
• Help shape the technical vision and long-term roadmap for the company's foundational AI systems.
• Own end-to-end architecture for complex AI systems, keeping them maintainable, observable and ethically aligned.
• Define and maintain SLOs for AI reliability, covering latency, uptime and model performance drift.
• Work with infrastructure engineers to improve GPU and TPU utilisation and speed up model inference pipelines.
• Lead cross-functional technical reviews of new AI deployments to confirm security, data governance and compliance standards are met.
• Mentor Principal and Senior Associates on scalable design, performance tuning and moving research into production.

Requirements

• You enjoy building systems, care about the quality of your work, and want to do the right thing. You want to solve problems that help change banking for good.
• You keep up with the latest research and can read scientific papers well enough to apply new techniques sensibly in production.
• You adapt quickly and bring structure to large, open-ended problems. You ask good questions, dig into root causes and explain findings clearly and concisely. You are willing to propose new ideas even when they are unproven.
• You have a deep technical background in engineering and mathematics. Your knowledge of hardware, software and AI helps you spot optimisation opportunities others miss.
• You can forge new paths toward business goals when the route is not clear.
• A bachelor's degree in computer science, AI, electrical or computer engineering, or a related field, plus at least 4 years of experience developing AI and ML algorithms or technologies. Alternatively, a master's degree in one of those fields plus at least 2 years of the same experience.
• At least 4 years of programming experience in Python, Go, Scala, CUDA or Java.
• Preferred: experience leading AI system development with trade-offs between cost, latency, throughput and accuracy.
• Preferred: 6+ years deploying scalable, responsible AI solutions on cloud platforms such as AWS, Google Cloud, Azure or an equivalent private cloud.
• Preferred: experience designing, delivering and supporting AI services.
• Preferred: experience building AI and ML technologies, such as LLM inference, similarity search and vector databases, guardrails or memory, in Python, C++, C, Java, CUDA or Golang.
• Preferred: experience applying state-of-the-art methods to optimise training and inference software for hardware utilisation, latency, throughput and cost.
• Preferred: experience building agentic AI systems and agentic workflows.
• Preferred: a habit of tracking the latest AI research and systems and applying new techniques in production with judgement.
• Preferred: experience designing distributed systems for model training, evaluation and online inference at petabyte scale.
• Preferred: experience defining AI model governance processes, including reproducibility, lineage tracking and automated retention schedules.
• Preferred: a track record of influencing architecture decisions across several AI product lines or platforms.

Conditions

The company may sponsor a new qualified applicant for employment authorisation for this role.

Full-time annual salary ranges by location: McLean, VA and Cambridge, MA: $197,300 to $225,100. New York, NY, San Francisco, CA and San Jose, CA: $215,200 to $245,600. Other locations follow their own range, and the exact offer amount appears in the offer letter. Part-time pay is prorated to the agreed hours.

The role is eligible for performance-based incentive pay, which may include cash bonuses and or long-term incentives (LTI). Incentives may be discretionary or non-discretionary depending on the plan.

Applications will be accepted for at least 5 business days.

No agencies. The company is an equal opportunity employer and considers applicants with criminal histories in line with applicable law.

Benefits

• Health, financial and other benefits. Eligibility depends on full- or part-time status, exempt or non-exempt status and management level.