Engenheiro(a) de Inteligência Artificial (Agentes Conversacionais e A2A)

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for an Engenheiro(a) de Inteligência Artificial (Agentes Conversacionais e A2A) based in Brazil.

This role offers the opportunity to engineer advanced AI agents capable of handling complex interactions and solving problems autonomously.

You will design conversational assistants, copilots, and multi-agent architectures that collaborate to achieve defined outcomes.

The position combines Python engineering with LLM orchestration, RAG, prompt engineering, and agentic AI.

You will work with modern technologies across the Google AI ecosystem, including Gemini and Vertex AI.

The role also involves connecting agents to APIs, databases, and external tools while ensuring reliable production performance.

You will establish quality metrics, improve model behavior, and build robust observability for AI applications at scale.

It is an environment suited to a technically strong professional who enjoys innovation, experimentation, and solving complex AI challenges.

Accountabilities:

• Design, develop, and continuously improve contextual, responsive, and human-centered conversational agents, including chatbots, copilots, and virtual assistants.
• Architect multi-agent and Agent-to-Agent (A2A) solutions in which autonomous agents communicate, delegate tasks, and collaborate toward shared objectives.
• Develop advanced prompt-engineering workflows and RAG pipelines to enable accurate use of proprietary knowledge bases while minimizing hallucinations.
• Integrate Google AI technologies and foundation models, particularly Gemini and Vertex AI, into production-ready applications.
• Implement tool-calling capabilities that allow agents to consume APIs, query databases, and execute actions across external and internal systems.
• Establish and monitor quality metrics covering conversational clarity, context retention, accuracy, and overall agent performance.
• Monitor inference health, application performance, scalability, and availability in production environments.
• Build and maintain robust APIs and backend services supporting AI applications and agent architectures.
• Implement observability, logging, and metrics solutions to identify issues and continuously improve reliability.
• Contribute to the evolution of AI engineering practices, architectures, and development standards.

Requirements:

• Completed higher education, preferably with a specialization in technology, artificial intelligence, software engineering, or a related field.
• Exceptional proficiency in Python, including object-oriented and asynchronous development.
• Extensive experience building and maintaining robust APIs using FastAPI, Flask, or Django.
• Advanced practical experience with LLM orchestration frameworks, particularly LangChain, LangGraph, and Langflow.
• Experience working with the Google AI ecosystem, especially Gemini and Vertex AI APIs and services.
• Strong understanding of agentic AI concepts, including autonomous reasoning patterns such as ReAct, Plan-and-Solve, and A2A workflows.
• Experience implementing and optimizing vector databases such as Qdrant, Pinecone, Milvus, Weaviate, or pgvector.
• Practical experience with production observability, logs, and metrics using tools such as Grafana.
• Experience with containerized environments and orchestration platforms such as Rancher.
• Strong analytical and problem-solving abilities, with a focus on building reliable, scalable AI solutions.
• Ability to work with autonomy, continuously learn emerging AI technologies, and translate experimentation into production-ready solutions.

Nice to have:

• Experience with LLM and RAG evaluation frameworks such as Ragas, TruLens, or LangSmith.
• Knowledge of Model Context Protocol (MCP) and standardized connections between agents and data sources.
• Experience implementing persistent long-term memory and context-management strategies for extended conversational sessions.
• Knowledge of AI application performance optimization, including latency reduction and token consumption management.

Benefits:

• Remote work model.
• Full-time employment.
• Working hours from 9:00 AM to 7:00 PM, Monday to Thursday, and 9:00 AM to 6:00 PM on Fridays.
• Opportunity to work on advanced applications of conversational AI, autonomous agents, RAG, and multi-agent architectures.
• Exposure to modern AI technologies and the Google AI ecosystem.
• Opportunity to contribute to innovative technology initiatives with meaningful real-world impact.
• Professional environment focused on continuous learning, technical development, and innovation.