Mid-Level AI Engineer — LLM Workflows & RAG (Hybrid)

Tamarind Intelligence; Barcelona; Hybrid; EUR 60,000 - 90,000; Full time

Benefits offered by this job

MacBook provided

Hybrid work model

Private health coverage

Pet-friendly office

Job summary

Biorce, Barcelona hub, is seeking a Mid-Level AI Engineer to advance our AI and Data engineering initiatives from our Barcelona hub. You will report to AI/Engineering leadership and contribute to building sophisticated AI systems beyond chat interfaces.

You will design agentic workflows, manage complex state in LLM applications, and ensure generative features are reliable, steerable, and scalable through strong software engineering practices.

Qualifications

• 2+ years in software/ML engineering with hands-on LLM deployments.
• Experience with orchestration tools (LangGraph, LangChain, CrewAI, Vertex AI Agent Builder).
• Experience with GCP ecosystem: Vertex AI Studio, Cloud Run, BigQuery, Cloud Storage.
• Advanced RAG techniques and hybrid search.
• Ability to manage conversational state and memory in complex AI graphs.

Responsibilities

• Design and implement multi-step AI workflows using LangGraph or similar frameworks.
• Develop, test, and version-control prompts (few-shot, Chain-of-Thought, ReAct).
• Use Vertex AI tools for prompt optimization and evaluation.
• Build and maintain Retrieval-Augmented Generation pipelines (parsing, embedding, vector DB).
• Establish automated evaluation frameworks to measure hallucination, groundedness, relevancy.

Skills

Software/ML engineering

LLM-based applications

Agentic mindset

State management in AI graphs

CI/CD and testing prompts

Tools

LangGraph

LangChain

CrewAI

Vertex AI Agent Builder

Vertex AI

BigQuery

Cloud Run

Job description

Biorce, Barcelona hub, is seeking a Mid-Level AI Engineer to advance our AI and Data engineering initiatives from our Barcelona hub. You will report to AI/Engineering leadership and contribute to building sophisticated AI systems beyond chat interfaces.

You will design agentic workflows, manage complex state in LLM applications, and ensure generative features are reliable, steerable, and scalable through strong software engineering practices.