The company is hiring a Senior Fullstack AI Engineer to design, build and ship production-grade retrieval-augmented generation (RAG) and agentic systems on Azure AI Foundry for our AI products.
You will own the end-to-end lifecycle of LLM features: ingestion and chunking pipelines, vector and hybrid search, prompt and agent orchestration, evaluation, and deployment on Azure. You will work closely with product, domain experts and client engineering teams, and set technical direction for less experienced engineers.<\/p>\n\n\n\n
Key responsibilities:<\/strong> Required qualifications:<\/strong> Preferred qualifications: Te ofrecemos<\/strong>
\u25cf Design and build RAG pipelines end to end: document ingestion, parsing (PDF, scanned documents, structured and semi-structured data), chunking strategies, embeddings, hybrid (vector + keyword) search and reranking.
\u25cf Build and operate on Azure AI Foundry: model deployments (Azure OpenAI and open weight models), prompt flows, Agent Service, Azure AI Search, evaluations and content safety.
\u25cf Implement agentic workflows (tool calling, multi-step planning, function orchestration) with frameworks such as Semantic Kernel, LangGraph or Autogen.
\u25cf Own evaluation: build offline and online eval harnesses (groundedness, relevance, faithfulness, latency, cost), golden datasets and regression gates in CI.
\u25cf Take systems to production: latency and cost optimization, caching, observability (Azure Monitor, tracing), rate limiting, failover across model deployments.
\u25cf Handle sensitive data correctly: PII redaction, compliance-aligned controls, Azure private networking, managed identity, Key Vault, RBAC.
\u25cf Partner with product managers and domain experts to turn ambiguous requirements into shippable features.
\u25cf Mentor engineers, review designs and code, and set standards for prompt engineering, evaluation and LLMOps across teams.
\u25cf Contribute to client-facing technical discussions, architecture reviews and estimates.
\u25cf Build the full feature surface, not just the model layer: backend APIs, data persistence, and the front-end chat, copilot or workflow screens that expose AI capabilities to users.<\/p>\n\n\n\n
\u25cf 8+ years of software engineering experience, with 3+ years of hands-on AI\/ML and LLM work, including systems you designed, shipped and supported in production (not prototypes or internal demos).
\u25cf Hands-on RAG experience: chunking, embedding models, vector databases (Azure AI Search, pgvector, Pinecone, Qdrant or similar), hybrid retrieval, reranking and citation\/grounding.
\u25cf Practical experience with Azure AI Foundry (formerly Azure AI Studio) and Azure OpenAI Service: model deployments, prompt flow, evaluations, content filters, quota and cost management.
\u25cf Strong Python; comfortable with FastAPI or similar service frameworks, async programming and containerization (Docker, AKS or Azure Container Apps).
\u25cf Experience with at least one orchestration framework: Semantic Kernel, LangChain\/LangGraph, LlamaIndex or Autogen.
\u25cf Demonstrated ability to evaluate LLM systems quantitatively and improve them iteratively.
\u25cf Solid grounding in software fundamentals: API design, data modeling, testing, CI\/CD (Azure DevOps or GitHub Actions), infrastructure as code (Bicep or Terraform).
\u25cf Experience working with sensitive data and security controls in regulated environments.
\u25cf Clear written and verbal communication; able to explain trade-offs to non-technical stakeholders.
\u25cf Proven full-stack engineer: has built and shipped complete web applications, owning front end, backend, data layer and deployment \u2014 not a backend or ML specialist with light UI exposure.
\u25cf Front end: expert in at least one modern framework (React, Angular, Vue or Blazor) with TypeScript, JavaScript or C#; responsive layouts, state management, streaming LLM responses (SSE or WebSockets), component testing and accessibility.
\u25cf Backend: strong in two or more of Python, Node.js, C#\/.NET or Java; REST and GraphQL API design, background jobs and queues (Azure Service Bus, Functions), authentication and authorization (Entra ID, OAuth 2.0, RBAC).
\u25cf Data: schema design and performance tuning on relational (PostgreSQL, Azure SQL) and document (Cosmos DB) stores; migrations, caching (Redis) and blob storage patterns.
\u25cf DevOps: containerized deployments (Docker, AKS or Container Apps), CI\/CD pipelines, infrastructure as code, secrets management, observability and on-call for systems you built.
\u25cf Owns features end to end \u2014 schema, API, UI, deployment, monitoring and support \u2014 and moves comfortably across the stack as the work demands.<\/p>\n\n\n\n
<\/strong>\u25cf Experience applying LLMs to document-heavy workflows in a regulated industry.
\u25cf Azure certifications: AI-102 (Azure AI Engineer Associate) or AZ-305 (Solutions Architect Expert).
\u25cf Fine-tuning or distillation of small models; experience with Azure Machine Learning or model catalog deployments.
\u25cf Multimodal RAG (images, tables, scanned documents) using Azure Document Intelligence.
\u25cf Experience with GraphRAG, knowledge graphs or structured retrieval over relational data.
\u25cf Familiarity with Model Context Protocol (MCP) and tool ecosystems for agents.
\u25cf Experience with Next.js or Blazor, design systems, and accessibility for AI-driven interfaces.
\u25cf Prior experience in a consulting or client-delivery setting.<\/p>\n\n\n\n
\u25cf Ownership of AI features that ship to real end users, not internal demos.
\u25cf A small, senior team with direct access to product leadership and clients.
\u25cf Modern Azure stack with budget for experimentation on new models and services.<\/p>\n\n\n\n