AI Engineer (LLM Agents & RAG)

Role overview Design and build production AI agents that perform real work for clients, including answering support questions from their own documents, qualifying leads, and screening candidates before escalating to a person. This is shipping engineering, not research: every agent goes to live users and has to remain reliable over time. The role sits inside a remote-first team serving clients across the US, UK, and India on a weekly delivery cadence.

Responsibilities - Build agents that answer from client data using retrieval, tool use, and clear guardrails. - Connect agents to systems clients already run, such as help desks, CRMs, and WhatsApp. - Build test sets from real conversations and measure quality before and after every change. - Monitor live agents, review transcripts, and improve answers over time. - Explain trade-offs to clients in plain language, including when a problem is not a fit for an LLM-based solution.

Requirements - Track record of shipping LLM-based features to production, including awareness of why similar features have failed. - Proficiency in TypeScript or Python and comfort working with APIs, queues, and databases. - Genuine care about evaluation and quality measurement, not just demos. - Ability to work in weekly increments and communicate progress clearly.

Nice to have - Experience with retrieval, embeddings, and vector or hybrid search. - Familiarity with Zendesk, Intercom, or the WhatsApp Business Platform.

Benefits and work setup - Remote-first team working with clients in the US, UK, and India. - Weekly cadence of shipping working software and showing it to clients. - Clients retain ownership of the source code and infrastructure built for them, so the next engagement is earned through delivery quality.