Senior Full-Stack AI Engineer – RAG & Intelligent Document Processing
About the role
Trigyn is seeking a Senior Full-Stack AI Engineer with deep expertise in cloud-native architecture, Retrieval-Augmented Generation (RAG), and Intelligent Document Processing (IDP) to support the development and implementation of the OSFG AI-Enabled Document Review Agent. This role is critical in designing, building, and deploying the core technical infrastructure, retrieval pipelines, and AI orchestration engines, ensuring that the solutions are scalable, evidence-grounded, resilient, and aligned with modern cloud and standards. The Senior Full-Stack AI Engineer will play a pivotal role in engineering the asynchronous review runtime, integrating Azure AI Search and Microsoft Foundry LLM services, building structure-aware document extraction pipelines, and enforcing strict citation validation to eliminate AI hallucinations.
Key responsibilities
Develop page-aware, structure-preserving extraction pipelines using Azure AI Document Intelligence for complex safeguard documents (PDF/DOCX) to enable precise page-level citations.
Architect and deploy the Review Orchestration Runtime using Azure Container Apps and Azure Service Bus to handle asynchronous review processing, rule selection snapshots, and status polling endpoints.
Design and configure dual search indexes in Azure AI Search (separating submitted document evidence from authoritative framework references) implementing hybrid keyword/vector retrieval and rule-scoped queries.
Build and fine-tune Microsoft Foundry prompt templates and structured JSON output schemas, ensuring evidence-grounded finding generation, missing-evidence fallbacks, and model/prompt version stamping.
Implement citation and schema validation mechanics to verify that every AI finding is validated against retrieved source text and framework clauses prior to persistence.
Design and maintain the Dataverse review data model (covering Review Runs, Rule Versions, AI Findings, Evidence, and Specialist Decisions) while enabling seamless integration with ISMS and Power Apps workflows.
Configure security, identity, and observability controls utilizing Microsoft Entra ID, managed identities, Key Vault, Application Insights, and Log Analytics.
Collaborate with cross-functional teams, including product owners and safeguard specialists, to architect, build, and test full-stack AI components and backend services.
About you
Bachelor's degree in Control and Automation Engineering, Computer Science, Information Technology, or a related engineering discipline.
Minimum 6 years of professional experience in full-stack AI and software engineering, with a strong track record of leading AI technical teams and engineering generative AI, NLP, computer vision, and large-scale data systems.
Deep expertise in RAG architecture design, vector search, hybrid retrieval, and enterprise search platforms utilizing Azure AI Search, Azure OpenAI, and custom embedding models.
Strong full-stack engineering proficiency in TypeScript, JavaScript, Python, Node.js/NestJS, and modern web frameworks (React).
Proven experience with cloud-native application development on Microsoft Azure, including Azure Container Apps, Service Bus, Key Vault, and Microsoft Foundry endpoints.
Extensive experience in Intelligent Document Processing (IDP), document layout parsing, structure-aware chunking, and text extraction.
Hands-on experience designing relational data models in Dataverse and integrating AI calls within Microsoft Power Platform environments (Power Apps, Power Automate).
Practical experience building automated evaluation harnesses, citation validators, structured output schema enforcement, and telemetry monitoring using Application Insights and Log Analytics.
Proven track record of delivering high-quality, scalable, and secure AI software solutions, including production platforms serving enterprise user bases.