As a Senior AI Engineer, you’ll build, scale and stabilize production-grade AI agent systems for a US tax technology company.

This is a hands-on senior role for someone who has moved beyond prototypes. You’ll design agentic workflows that connect with backend services, frontend events and internal tools, and you’ll own AI services end to end: Python APIs, agent orchestration, AWS deployment, monitoring and continuous improvement, in an environment with sensitive data and strict operational constraints. 100% remote from Latin America, full-time contracting.

What you’ll do

• Design, build and optimize AI-driven systems, agentic workflows and tool-using agents.

• Build and maintain MCP servers or similar tool-calling layers so agents interact safely with backend APIs and business logic.

• Develop AI services and APIs with Python and FastAPI.

• Deploy, operate and scale AI applications on AWS (Bedrock, Agent Core, Guardrails, ECS, ECR, Lambda, CloudWatch).

• Implement security, validation, guardrails and prompt-injection mitigation for LLM applications handling sensitive data.

• Debug, refactor and stabilize existing production LLM services to improve reliability, latency and maintainability.

• Work with frontend teams (React, TypeScript) to bring AI agents into user-facing product experiences.

• Ship through CI/CD with GitHub Actions, Docker and Terraform.

• Communicate trade-offs, risks and architectural decisions to technical and non-technical stakeholders.

What you’ll bring

• 5+ years of professional software engineering, with a track record of deploying and maintaining production systems.

• Strong Python.

• Hands-on experience building AI agents or agentic workflows in production (LangChain, LangGraph, LlamaIndex, CrewAI, Strands Agents or custom frameworks). Prototype-only experience is not enough.

• Experience with MCP servers, tool calling or function calling.

• Hands-on LLM development with OpenAI, Anthropic, AWS Bedrock or similar.

• APIs and microservices with FastAPI or an equivalent framework.

• Deploying, operating and monitoring services on AWS.

• Docker, CI/CD (GitHub Actions) and Terraform or similar IaC.

• Solid grasp of production reliability, observability and debugging.

• C1 English or higher.

Nice to have

Vector databases, semantic search, AWS Textract, AWS ECS, pytest.

Why this role

You’ll take AI agents from promising demos to systems people rely on, in a product where correctness and security are not optional.