AI Agent Engineer — Build Proactive LLM Systems

TextYess; España; Presencial; EUR 60.000 - 95.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Ownership from day one

Lean team with international talent

High-impact projects

Rapid growth opportunities

Collaborative culture

Descripción de la vacante

TextYess in Spain is hiring for its AI team to build the agentic runtime for proactive agents: context building, tool orchestration, scoring, and guardrails. You’ll own production components from day one and work directly on the signal vs. noise call.

You’ll address grounding errors, balancing exact-match and semantic search, ensure multilingual quality, and reduce latency and cost while shipping measurable improvements using Langfuse traces and evals.

Formación

• 2+ years shipping software professionally, including at least one LLM feature touch — a chatbot, a RAG system, or an agent.
• You know how RAG actually behaves: embeddings, vector search, chunking, reranking, and where each one breaks.
• You can read traces and diagnose production issues: retrieval misses, prompt problems, or misused tools.
• Strong Python and backend fundamentals, with a bias for measuring latency, cost and quality.
• Familiarity with multilingual customer interactions is a plus.

Responsabilidades

• Build and own production components from day one for proactive AI agents.
• Address grounding: fix product availability and ranking mismatches.
• Balance exact-match vs semantic search in a single ranking system.
• Ensure multilingual quality across languages.
• Profile requests, test prompts, and ship latency/cost improvements.
• Design evals to catch problems before customers see them.

Conocimientos

Python

Backend fundamentals

LLM features

Embeddings / vector search

Latency optimization

Langfuse / LangSmith

Weaviate / Elastic / Algolia

Herramientas

Langfuse

LangSmith

Braintrust

Arize

Humanloop

Elastic

Algolia

Qdrant

Weaviate

Descripción del empleo

TextYess in Spain is hiring for its AI team to build the agentic runtime for proactive agents: context building, tool orchestration, scoring, and guardrails. You’ll own production components from day one and work directly on the signal vs. noise call.

You’ll address grounding errors, balancing exact-match and semantic search, ensure multilingual quality, and reduce latency and cost while shipping measurable improvements using Langfuse traces and evals.