AI Engineer – Core (Agentic AI Systems)

About the role

Hilbert is building a reasoning engine that transforms months‑long decision cycles into minutes by orchestrating multi‑step inference over messy enterprise data. As an AI Engineer – Core you will own critical pieces of the AI stack, designing agent architectures, building evaluation systems, and balancing accuracy, latency and cost in production.

Key responsibilities

• Design and implement agent workflows using LangChain, LangGraph or equivalent frameworks, including state memory, routing and tool registries.
• Own the evaluation layer for agents: harnesses, metrics, golden datasets, regression gates and human‑in‑the‑loop review.
• Operate production systems end‑to‑end: tracing, monitoring, latency budgeting, on‑call support and cost‑per‑task optimization.
• Diagnose and fix production failures such as hallucinations, tool misuse, retrieval misses and silent degradation.
• Set technical standards for agent work, review designs and define reusable patterns for the team.

Required profile

• 4+ years of production software engineering experience with APIs, services and data infrastructure.
• 2+ years building LLM‑based or agentic systems that are used by real users in production.
• Hands‑on experience with LangChain, LangGraph or comparable orchestration frameworks.
• Strong communication skills to explain technical decisions to non‑technical stakeholders.
• Ability to work autonomously, take ownership and thrive in ambiguous, fast‑moving environments.

Required skills

• LangChain
• LangGraph
• Retrieval‑augmented generation (RAG) – hybrid and graph‑based retrieval
• Observability tools such as Langfuse or OpenTelemetry
• Tool‑calling frameworks, structured output and constrained decoding

What we offer

• Competitive salary plus equity package, aligned with experience.
• Performance‑based bonuses tied to project milestones and customer impact.
• Fast‑growing startup environment with real ownership and impact.