Senior AI Engineer — Production LLM & Agent Systems

Hilbert's AI; Turkey; On-site; TRY 900,000 - 1,200,000; Full time

Job summary

Hilbert’s AI is seeking an AI Engineer to build production-grade AI systems end-to-end—from prototype to pipeline to product—with startup-pace ownership. You will design agent architectures, implement evaluation systems, and balance accuracy, latency, and cost in production.

You’ll lead evaluation of agents, build regression gates, and ensure scalable, auditable deployments across the AI stack. Strong emphasis on shipping robust, observable systems in a fast-moving environment.

Qualifications

• 4+ years in production software engineering (APIs, services, data infra) with tests and CI/CD.
• 2+ years building LLM or agent systems in production.
• Hands-on experience with LangChain, LangGraph, or equivalent frameworks.
• Ability to explain technical decisions clearly to non-technical stakeholders.
• Takes ownership, ships, and communicates effectively.

Responsibilities

• Own evaluation layer for agents — harnesses, metrics, regression gates, and reviews.
• Architect and implement agent workflows using LangChain/LangGraph; memory, routing, tools.
• Own end-to-end systems from experimentation to production including monitoring and cost.
• Diagnose production failures and turn them into durable fixes and tests.
• Set technical standards for agent work and raise team capabilities.
• Collaborate with founding team and cross-functional partners to communicate tradeoffs.

Skills

Production software

LLM/agent systems

LangChain

LangGraph

Clear communication

Ownership

Ambiguity

Startup speed

Tools

LangChain

LangGraph

Job description

Hilbert’s AI is seeking an AI Engineer to build production-grade AI systems end-to-end—from prototype to pipeline to product—with startup-pace ownership. You will design agent architectures, implement evaluation systems, and balance accuracy, latency, and cost in production.

You’ll lead evaluation of agents, build regression gates, and ensure scalable, auditable deployments across the AI stack. Strong emphasis on shipping robust, observable systems in a fast-moving environment.