Remote job
Senior AI Engineer
Job details
About this role
Role overview Build and operate AI features that help commercial bankers understand what is changing in a client’s financial data and decide what to do next. This is a senior, hands-on engineering role: you’ll shape how models reason over structured information, connect answers to evidence, and recognize when they should abstain. You’ll own work from design through production and help determine how the assistant evolves.
Responsibilities - Design the model reasoning layer, including context, tool selection, and multi-step handling of questions, with answers traceable to source data. - Build evaluation datasets and test harnesses to measure whether changes improve answer quality; turn user corrections into new tests. - Select models and manage the trade-offs among answer quality, reliability, latency, and cost, using conventional statistical methods where they fit better. - Work with data engineers on semantic definitions and signals that identify meaningful changes in client information. - Develop agent tools and Model Context Protocol servers, and add tracing to understand failures and answer behavior. - Ship, monitor, and maintain production services, using evidence to explain technical decisions and guide improvements.
Requirements - Experience shipping LLM features used in practice, designing model interactions with tools or structured data, and supporting them when issues arise. - Hands-on experience building evaluations and analyzing model behavior. - Strong quantitative judgment and the ability to assess data in context. - Strong Python or another general-purpose programming language, SQL, and sound service engineering practices, including tested, modular code. - Some experience with CI/CD, Docker, cloud services, and tracing for LLM systems. - Ability to lead technical work through evidence and explain trade-offs to nontechnical partners.
Nice to have Experience in fintech, lending, or accounting; Spark or Databricks; semantic layers or metric views; or model fine-tuning.
Benefits and work setup Hybrid or fully remote, with no required office schedule. The team emphasizes end-to-end ownership, shipping useful improvements, and sustainable working hours.