Remote AI Engineer (LLM Systems & Automation)

HireLatam is a premier recruitment agency that places top Latin American talent in independent contractor roles in US companies. With a proven track record and a commitment to excellence, we're your trusted partner in the pursuit of career success. Our extensive network, personalized approach, and supportive guidance ensure that you're in the best hands to find your next job opportunity. Job Title: Remote AI Automation Engineer (LLM Systems & Automation) (100% Work From Home) Location: Remote from Latin America Position Type: Full-time Salary: Up to $3,000 USD/month (depending on experience and technical assessment) Schedule: Schedule flexibility based on project needs; aligned with core leadership hours. Job Overview Our Client is a forward-thinking digital media and technology organization dedicated to leveraging cutting-edge automation, artificial intelligence, and custom software solutions to drive business performance. Operating with a lean, agile structure, our client focuses on converting complex operational challenges and strategic goals into high-impact, reliable tools. By combining low-code automation platforms with custom development and LLM capabilities, they foster a culture of resourcefulness, rapid deployment, and practical execution. They are looking for a hands-on engineer to organize and improve our existing AI tools and workflows. You'll structure data, manage what information each model receives, choose models for different tasks, and test results for accuracy, speed, and cost. You'll work directly with the founders and help the team use these tools. The Role We're looking for a hands-on engineer to organize and improve our existing AI tools and workflows. You'll structure data, manage what information each model receives, choose models for different tasks, and test results for accuracy, speed, and cost. You'll work directly with the founders and help the team use these tools. Our Client is seeking a hands-on engineer to organize and improve their existing AI tools and workflows. You'll structure data, manage what information each model receives, choose models for different tasks, and test results for accuracy, speed, and cost. You'll work directly with the founders and help the team use these tools. Working directly alongside the company founder, this professional will evaluate technical requirements, select the optimal stack (combining automation platforms, custom code, and LLM integrations), and take full ownership of building, testing, and deploying end-to-end solutions. Key Responsibilities Workflow Automation & System Integration Builds, tests, troubleshoots, and maintains scalable workflows and automations using n8n. Integrates REST/GraphQL APIs, webhooks, relational/non-relational databases, and diverse third-party platforms. Evaluates technical trade-offs to choose the simplest, most effective architecture—balancing low-code automation with custom script development. Custom Development & AI/LLM Implementation Writes clean, efficient code in JavaScript/TypeScript or Python when custom logic or data processing is required. Implements LLM capabilities and AI model integrations directly into core business operations, reporting systems, and research tools. Uses AI-assisted coding tools to accelerate development while independently reviewing, debugging, validating, and testing all generated code for accuracy and security. Solution Architecture, Testing & Delivery Translates broad business targets into incremental technical releases, validating core assumptions early in the development cycle. Conducts rigorous quality assurance to test solutions against edge cases, common failure modes, data integrity risks, and usability bottlenecks. Manages client credentials, sensitive data, and environment configurations in strict accordance with security best practices. Produces clear, concise system documentation to ensure solutions remain maintainable and operational long after initial deployment. Central Responsabilities of the role Structure data. Turn documents, conversations, and other source material into consistent, usable data, preserving source references and client access boundaries. Manage context. Decide what instructions, information, tools, and previous work each model needs at each step. Preserve useful context across longer projects and reduce unnecessary processing. This is often called context engineering. Choose models and organize the work. Decide which model handles each task, the order of the steps, and the handoffs between them. Use lower-cost models where tests show they meet the quality standard. Measure performance. Track accuracy, errors, token usage, subscription limits, cost, and completion time. Find where our systems waste resources or produce unreliable results. Keep improving the setup. Build repeatable tests for our actual tasks. Compare new models with the current setup and