Applied AI Engineer (LLM)

Role overview Join a small AI & data group building language-model features that real businesses rely on, moving past prototype demos toward dependable, client-ready systems. The work centers on retrieval, grounded assistants, and agentic workflows where humans stay in the loop on consequential actions.

Responsibilities - Design and ship LLM-powered features such as retrieval-augmented generation, grounded assistants, and agentic workflows tied to real customer data. - Build evaluation harnesses and guardrails so model output is reliable enough to put in front of clients. - Define where humans must remain involved, including clear boundaries around irreversible actions. - Monitor and control the tradeoffs between latency, cost, and quality so features are actually shippable. - Partner with product and clients to identify high-value AI use cases and push back when AI is not the answer.

Requirements - 4+ years of software engineering experience, including production work with LLMs or ML APIs. - Strong Python and/or TypeScript, plus comfort with the modern AI tooling ecosystem. - A skeptical mindset about model output, with a habit of building checks for known failure modes. - Clear communication skills, including explaining AI capabilities and limits to clients or non-technical stakeholders.

Nice to have - Experience with retrieval, vector search, or grounding models on private data. - Familiarity with prompt evaluation, fine-tuning, or building agentic systems safely. - A side project or demo showing an AI idea taken end to end to something usable.

Benefits and work setup - Fully remote across EU and Americas timezones, async-friendly with real overlap hours. - Hardware and API or compute budget, a yearly learning stipend, and meaningful equity.