Senior Full Stack Developer Agentic AI positions focus on delivering results in their domain. This page aggregates open Senior Full Stack Developer Agentic AI roles and what employers typically expect.
At EX Squared LATAM, we are a digital transformation company focused on building innovative, high-impact technology solutions that drive real business value. We partner with top-tier clients across global, regional, and local markets, fostering a collaborative, inclusive, and innovation-driven culture where your ideas truly matter. Role Overview We’re currently looking for a Senior Full Stack Developer with strong experience in React, Node.js, TypeScript, and agentic AI workflows. You’ll help our client build and maintain a catalog of AI agents designed to support the software development lifecycle, including code generation, quality assurance, product management, design-to-code, and content-authoring workflows. This position combines full-stack engineering with prompt engineering, agent development, evaluation, observability, and continuous experimentation. The ideal candidate is a collaborative and proactive engineer who can explore emerging AI technologies, turn ideas into practical solutions, gather feedback from delivery teams, and continuously improve agent performance. Location: Fully remote position within LATAM. Contract Duration: 6-month fixed-term contract with possibilities of extension. Language Requirement: Advanced English. What you'll do - Build and maintain AI agents supporting software development, QA, and product teams. - Develop code-generation, coding-standard, design-to-code, and workflow automation agents. - Design structured, dependency-aware agentic workflows rather than single-prompt solutions. - Create agent skills, prompts, tool interfaces, and reusable workflow components. - Implement evaluation and observability practices to monitor agent quality and performance. - Develop self-validation and error-recovery loops using testing and type-checking results. - Build and maintain applications using React, Node.js, and TypeScript. - Develop and consume REST and GraphQL APIs using established best practices. - Integrate agents with internal systems, SaaS platforms, APIs, and development tools. - Collaborate with delivery teams, collect feedback, and improve existing agent solutions. - Research emerging AI tools, frameworks, vendors, and local model alternatives. - Document agent behavior, architecture decisions, limitations, and usage guidelines. - Apply security, accessibility, testing, and performance standards throughout development. - Participate actively in team discussions, technical planning, and solution design. What you'll bring - 5+ years of professional full-stack development experience. - Strong hands-on experience with React, Node.js, and TypeScript. - Experience building, integrating, or maintaining agents and agentic workflows. - Knowledge of prompt engineering, task decomposition, context management, and tool calling. - Understanding of agent evaluation, observability, validation, and error-recovery patterns. - Experience consuming GraphQL APIs, ideally using Apollo Client. - Experience with REST APIs and third-party SaaS integrations. - Familiarity with AWS, Docker, and cloud-based application architecture. - Experience with automated testing tools such as Jest, Cypress, or Vitest. - Understanding of secure API development, OAuth2/OIDC, API keys, and access controls. - Experience developing accessible interfaces aligned with WCAG 2.1 AA. - Ability to structure clean, scalable, and maintainable application code. - Strong documentation and technical communication skills. - Advanced English and the ability to collaborate in a distributed environment. - A proactive, exploratory mindset with the ability to work independently. - A Mac computer meeting the client’s BYOD requirements, preferably with 32 GB RAM. What will make you stand out - Experience creating agent skills using Markdown-based instructions. - Knowledge of RAG and other retrieval-based AI patterns. - Experience with LLM structured outputs, function calling, or tool calling. - Familiarity with golden test sets, LLM-as-judge, and…