AI Engineer (LLM/Agents)

Role overview

A fully remote engineering position focused on building a new AI-powered analytics product that interprets financial markets and global news for investors. The role centers on designing LLM-driven features and agentic workflows that turn raw information into grounded, decision-ready narratives. It sits within a remote-first, small-team environment with high autonomy and flexible scheduling.

Responsibilities

- Design and ship LLM-powered features for market and news understanding, including summarization, extraction, classification, and contextual investor narratives. - Build multi-step agentic workflows that gather evidence, run checks, and produce reliable, grounded outputs. - Own model selection and routing, balancing quality, latency, and cost with fallbacks and safe degradation. - Implement tool and function-calling patterns connecting LLMs to internal data services and external signals, including schema design, constraints, and error handling. - Develop evaluation pipelines combining offline golden sets, regression tests, and online A/B experiments tied to product outcomes. - Optimize token usage, context strategy, caching, and verbosity control against cost and performance targets.

Requirements

- Strong software engineering fundamentals with a track record of shipping production systems, not demos. - Practical background integrating LLMs into real products rather than casual experimentation. - Solid grasp of agent patterns and tool use. - Ability to write clear specifications and guidelines that both engineers and models can follow. - Comfort owning ambiguous problems end-to-end from prototype through monitoring and iteration.

Nice to have

- Genuine curiosity about finance and markets; no deep finance background required. - Familiarity with adjacent parts of the stack including Python services, APIs, queues, caching, and observability tooling, plus React, TypeScript, .NET Core, C#, SQL Server, and Node.js for cross-team collaboration.