Senior AI Engineer: End-to-End LLM Systems in Production
Posted Updated
Description
Jeeves, the stablecoin-native banking platform, is hiring an AI Engineer to own end-to-end AI workflows, from design through production. You'll work with Product and Backend Engineering to build and ship AI-powered systems, with a focus on reliability and measurable impact.
This senior, hands-on role requires crafting agentic architectures and strong Python skills. You will design multi-step workflows, manage costs and latency, implement observability, and maintain audit trails for regulated
Design agentic and LLM-based systems for multi-step work with human-in-the-loop points. Write design docs, make build-vs-buy and model choices, set success metrics. Build production-grade LLM pipelines: prompt design, context management, validation, fallbacks, and confidence scoring. Integrate AI services with Jeeves' backend: clear API contracts, retries, graceful degradation, and per-customer data isolation. Manage cost and latency across models and select appropriate models for each step. Build evaluation sets and automated evals for every workflow you own; catch regressions when prompts or data change. Instrument AI components with logging, tracing, dashboards, and alerts; maintain audit trails for regulated financial products. Set patterns and shared tooling so other engineers can build on your work; mentor the team on coding agents and AI tooling. 7+ years of professional software engineering experience. 2+ years building and operating LLM/AI-powered production systems. Own large end-to-end AI workflows from scoping and design through launch and iteration. Hands-on experience shipping LLM-powered applications with APIs and structured outputs. Experience designing agentic or multi-step AI workflows with tool use and human review steps. Strong proficiency in Python and backend fundamentals (APIs, databases, async, cloud). Professional fluency in English, written and spoken.Skills
Sign in to read and join the discussion.