Senior AI Architect – Multi-Agent LLM Systems

Metova Inc; Ciudad de México; On-site; MXN 1,000,000 - 1,400,000; Full time

Job summary

Metova Inc is seeking an AI Engineer in Mexico City to design and supervise architectures for intelligent agents and LLM-based solutions. You will integrate LangChain, LlamaIndex, AutoGen, and CrewAI, driving multi-agent coordination and autonomous workflows across business contexts.

You will collaborate with product, UX, data, and backend teams to translate business needs into scalable agent architectures, and work with MLOps to deploy updates while managing memory and long-term planning.

Qualifications

• 5 years of experience in AI projects.
• 2 years in implementing autonomous agents or co-pilots.
• Fluent technical English.
• Experience with business data in accounting/finance/payroll/ERP is a plus.
• Experience with vector stores and RAG architectures.

Responsibilities

• Define, design, and supervise the technical architecture of solutions based on intelligent agents and LLMs, integrating tools such as LangChain, LlamaIndex, AutoGen, CrewAI, or equivalent frameworks.
• Implement MCP (Model Context Protocol) and A2A (Agent-to-Agent) architectures to enable multi-agent coordination and autonomous flows within business environments.
• Work with the MLOps team and execution environments that enable continuous agent updating and deployment, including memory management, context, and long-term planning.
• Collaborate closely with product, UX, data, and backend teams to map business needs to intelligent agent architectures.

Skills

Autonomous agent design

Tools

LangChain

LlamaIndex

AutoGen

CrewAI

Python

FastAPI

Pydantic

Job description

Metova Inc is seeking an AI Engineer in Mexico City to design and supervise architectures for intelligent agents and LLM-based solutions. You will integrate LangChain, LlamaIndex, AutoGen, and CrewAI, driving multi-agent coordination and autonomous workflows across business contexts.

You will collaborate with product, UX, data, and backend teams to translate business needs into scalable agent architectures, and work with MLOps to deploy updates while managing memory and long-term planning.