AI & Agent Engineer: Graph-RAG & LLM Architect

Appnovation; Medellín; Presencial; COP 40.000.000 - 70.000.000; Jornada completa

Descripción de la vacante

Appnovation is seeking an AI / Agent Engineer to join our Medellín tech team. You will build and evolve the agentic layer of our platform, crafting MCP servers, tools and graph-RAG pipelines that enable documents to be understood by multiple LLMs.

You will partner with product, data and QA to raise answer quality and grounding. The role requires 5+ years in software engineering with a focus on AI/ML or LLM-based apps, strong Python and experience with MCP servers, RAG systems, and graph

Formación

• Bachelor's degree or equivalent practical experience.
• 5+ years of software engineering with 2+ years in AI/ML or LLM-based apps.
• Strong Python and hands-on experience in LLM development and agentic patterns.

Responsabilidades

• Build and evolve MCP servers, tools, and skills for multi-LLM exposure.
• Design graph-RAG pipelines transforming documents into agent-friendly facts.
• Implement retrieval across tiers (source, vector-RAG, graph) and temporal replay.
• Collaborate with QA to improve answer quality and grounding.
• Write clean, tested production-grade code and participate in reviews.
• Work with product, data, and QA in an agile/sprint model.
• Document designs, prompts, and tool interfaces for maintainability.
• Monitor deployed agents/models and iterate on performance.

Conocimientos

Python

LLM development

AI/ML

MCP servers

RAG systems

Educación

Bachelor's Degree in CS

Herramientas

Neo4j

Postgres

Git

CI/CD

VectorDBs

Descripción del empleo

Appnovation is seeking an AI / Agent Engineer to join our Medellín tech team. You will build and evolve the agentic layer of our platform, crafting MCP servers, tools and graph-RAG pipelines that enable documents to be understood by multiple LLMs.

You will partner with product, data and QA to raise answer quality and grounding. The role requires 5+ years in software engineering with a focus on AI/ML or LLM-based apps, strong Python and experience with MCP servers, RAG systems, and graph