Principal AI Engineer

This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a Principal AI Engineer based in Brazil. This role leads the architecture and development of enterprise-scale AI orchestration and retrieval systems. You will define how LLMs connect with tools, data sources, APIs, and specialized agents. The position combines strategic technical leadership with hands-on engineering across agentic workflows and RAG systems. You will establish standards for MCP servers, tool exposure, retrieval, context management, evaluation, and observability. Working across teams, you will create reusable abstraction layers that enable product engineers to build effectively on top of AI capabilities. This is an opportunity to shape the technical foundations of a central AI platform while mentoring engineers and driving best practices. Accountabilities: - Design and develop orchestration and abstraction layers connecting LLMs with tools, data, and specialized sub-agents. - Build and operate MCP servers while establishing standards for defining, exposing, and versioning tools. - Define tool-surface strategies, including the appropriate number of tools and APIs exposed to LLMs and how they remain coherent and discoverable. - Establish when to use specialized sub-agents versus direct tool exposure and develop corresponding multi-agent patterns. - Design and optimize RAG systems covering chunking, embeddings, vector stores, hybrid search, re-ranking, and context assembly. - Create abstraction layers that decouple product teams from underlying models, tools, and providers. - Develop routing, memory, and context-window management strategies for agentic workflows. - Define evaluation frameworks covering retrieval quality, tool-selection accuracy, task success, latency, and cost. - Establish observability and tracing across multi-step agent and tool interactions. - Implement safety measures, guardrails, authentication, and access controls across AI tools and agents. - Collaborate with product teams to onboard new capabilities as tools and agents into the central AI system.

- Mentor engineers and help raise orchestration, retrieval, and agentic AI engineering standards across teams. Requirements: - Advanced English proficiency at C1/C2 level for fluent communication across distributed teams. - 10+ years of experience in software or AI engineering, including hands-on development of LLM orchestration, agents, and retrieval systems. - 4–5+ years of deep hands-on experience with LLM orchestration frameworks such as LangGraph, LlamaIndex, Semantic Kernel, or equivalent technologies. - Direct experience building MCP servers and implementing tool or function-calling integrations. - Evidence-based expertise in designing effective LLM tool surfaces, including decisions around the number of tools and APIs exposed through MCP servers. - Strong understanding of the tradeoffs between specialized sub-agents and direct tool exposure, with experience designing multi-agent architectures. - Deep expertise in retrieval and RAG, including chunking strategies, embeddings, vector databases, hybrid or keyword search, and re-ranking. - Experience designing abstraction layers and platform APIs used by multiple engineering teams. - Strong knowledge of context-window management, prompt and context assembly, and AI cost and latency optimization. - Experience implementing evaluation, monitoring, observability, and tracing for agentic and retrieval systems. - Ability to establish technical strategy and standards while remaining actively involved in coding and architecture. - Strong stakeholder management, collaboration, communication, and problem-solving skills. Benefits: - Competitive salary paid in USD. - 100% remote work. - Independent Contractor agreement. - Long-term contract. - 2 weeks of paid time off (PTO). - Holidays based on the North American calendar.