AI DevOps

act digital; Lisboa; Híbrido; EUR 70 000 - 100 000; Tempo integral

Resumo da oferta

act digital in Lisbon, Braga or Coimbra is seeking an AI DevOps Engineer to own production-grade Generative AI solutions and help scale them across the business.

You will work with Azure OpenAI, Azure AI Foundry, and related services, bridging software development with platform operations. The role requires strong DevOps, GenAI and cloud experience and offers hybrid work arrangements.

Qualificações

• 5–8 years of software development experience ideally, including 2+ years building/operating GenAI or LLM solutions in production.
• Strong Python and/or TypeScript, with solid CI/CD, IaC (Bicep/Terraform) and containerization (Docker) skills.
• Hands-on experience with Azure cloud (App Service, Functions, AKS), Azure OpenAI, Azure AI Foundry and REST API integrations.
• Practical experience with at least three GenAI patterns: conversational agents, agentic workflows, RAG, document intelligence, or AI-driven integrations.
• Experience taking over and stabilizing vendor-built codebases, plus backlog management with business stakeholders.

Responsabilidades

• Take over and stabilize AI solutions built by external vendors (agents, copilots, RAG apps, document intelligence, LLM integrations), reviewing code, prompts and architecture to close technical debt.
• Design and ship new AI features using Azure OpenAI, Azure AI Foundry, Copilot Studio and Azure AI Services, applying RAG, agentic workflows and fine-tuning as appropriate.
• Build and maintain CI/CD pipelines (GitHub Actions/Azure DevOps) and apply LLMOps practices: prompt versioning, evaluation, A/B testing and regression testing.
• Implement observability for cost, usage and response/grounding quality, and act as L3 support for incident resolution and permanent fixes.
• Manage AI backlog with business stakeholders, translating needs into user stories and driving adoption.

Conhecimentos

Python

TypeScript

CI/CD

IaC

Docker

Azure

GenAI patterns

Ferramentas

GitHub Actions

Azure DevOps

Descrição da oferta de emprego

Join an international project as an AI DevOps Engineer, taking ownership of production-grade Generative AI solutions and helping scale them across the business. This role is ideal for a DevOps professional who has moved into AI/LLM solutions, or an AI engineer with strong DevOps and platform operations skills.

Location: Lisbon, Braga or Coimbra (hybrid 2x per week in one of these locations) — candidates must be already based in Portugal

Responsibilities

• Take over and stabilize AI solutions built by external vendors (agents, copilots, RAG apps, document intelligence, LLM integrations), reviewing code, prompts and architecture to close technical debt.
• Design and ship new AI features using Azure OpenAI, Azure AI Foundry, Copilot Studio and Azure AI Services, applying RAG, agentic workflows and fine-tuning as appropriate.
• Build and maintain CI/CD pipelines (GitHub Actions/Azure DevOps) and apply LLMOps practices: prompt versioning, evaluation, A/B testing and regression testing.
• Implement observability for cost, usage and response/grounding quality, and act as L3 support for incident resolution and permanent fixes.
• Manage AI backlog with business stakeholders, translating needs into user stories and driving adoption.

Requirements

• 5-8 years of software development experience ideally, including 2+ years building/operating GenAI or LLM solutions in production.
• Strong Python and/or TypeScript, with solid CI/CD, IaC (Bicep/Terraform) and containerization (Docker) skills.
• Hands-on experience with Azure cloud (App Service, Functions, AKS), Azure OpenAI, Azure AI Foundry and REST API integrations.
• Practical experience with at least three GenAI patterns: conversational agents, agentic workflows, RAG, document intelligence, or AI-driven integrations.
• Experience taking over and stabilizing vendor-built codebases, plus backlog management with business stakeholders.

Nice to have

• Agentic frameworks (LangGraph, Microsoft Agent Framework, Pydantic AI).