Staff AI Engineer - Agentic Shopping Leader

Relocate.me; Amsterdam; On-site; EUR 120,000 - 190,000; Full time

Benefits offered by this job

Relocation package

Flight tickets

Temporary housing

Housing search assistance

Visa services

Adaptation tips

Job summary

Picnic is seeking a Staff AI Engineer to lead the Agentic Shopping initiative, building a high-scale conversational AI stack that understands households, assembles baskets, and learns from user interactions. You will own technical direction for multi-turn models, retrieval, and production systems, balancing cost, latency, and reliability.

You will drive RAG and context-management strategies, evaluate safety, and mentor engineers across teams.

Qualifications

• Proven experience in multi-turn conversational AI architectures and production systems.
• Experience training and optimizing open-weight LLMs for production deployment.
• Hands-on experience building production-grade RAG systems and context management.
• Strong ability to design robust evaluation and monitoring loops with safety guardrails.
• Ability to lead technical direction and mentor engineers without formal hierarchy.
• Strong Python skills and experience with PyTorch and related tooling.

Responsibilities

• Lead technical direction for agentic shopping experiences.
• Architect multi-turn conversational system with latency and safety constraints.
• Prototype and productionize AI components, from model to deployment.
• Collaborate with product and design to align on capabilities and UX.
• Define evaluation, observability, and fallback patterns.

Skills

Conversational AI

LLMs fine-tuning

RAG systems

Evaluation loops

Technical leadership

Python

PyTorch

LangGraph

vLLM

Pydantic AI

Qdrant

DeepEval

Education

Degree in Computer Science or AI

Tools

PyTorch

LangGraph

vLLM

Pydantic AI

Qdrant

DeepEval

Job description

Picnic is seeking a Staff AI Engineer to lead the Agentic Shopping initiative, building a high-scale conversational AI stack that understands households, assembles baskets, and learns from user interactions. You will own technical direction for multi-turn models, retrieval, and production systems, balancing cost, latency, and reliability.

You will drive RAG and context-management strategies, evaluate safety, and mentor engineers across teams.