Wanted: Senior AI Architect, GenAI & Document Intelligence

Company Description Alyra Technology builds AI-enabled software and automation systems for document-heavy, regulated industries: insurance, legal, finance and healthcare. We pair modern Microsoft Azure architecture with applied generative AI to ship systems that are explainable, auditable and production-ready, not demos. Our work centers on pulling structure, meaning and decisions out of complex, unstructured documents at scale. Role Description We're hiring a senior Azure architect who's designed production GenAI systems for messy, real-world documents, and done it for years. Contracts, filings, policies, medical records, financial statements. If you've bolted GPT onto a web app, this isn't your role. If you've built production GenAI pipelines for document-heavy, regulated work, keep reading. This role is architecture-first. You'll turn ambiguous AI ideas into clear, defensible system designs covering data flows, agent interactions, retrieval pipelines, security boundaries and auditability, primarily on Microsoft Azure. We pull you in as projects launch, so the cadence flexes with the work, lighter in steady stretches and heavier during active builds. Qualifications 5+ years designing and shipping production AI systems, not just prototyping them 5+ years working directly with unstructured text and document-centric AI. This is non-negotiable. Proven production RAG, multi-agent or document intelligence pipelines at real-world complexity Deep, hands-on Azure platform experience, end to end A real understanding of how GenAI fails in production: hallucination, retrieval quality, context limits, latency and cost at scale Ability to design for auditability and explainability in compliance-sensitive environments 7+ years total software or platform engineering experience Located in Brazil, with professional English, written and spoken Preferred Qualifications Hands-on with complex multi-page documents: contracts, legal filings, medical records, financial statements, insurance forms or regulatory submissions Layout-aware parsing tools such as Azure Document Intelligence, Unstructured or LayoutLM Document-grounded output systems where every answer traces back to a specific source passage or page Client-facing architecture experience in insurance, legal, finance or healthcare Comfort working alongside AI-assisted development tools such as Claude Code How To Apply Submit your resume in English.