AI Platform Engineer - Syone

Syone; Lisboa; Presencial; EUR 90 000 - 130 000; Tempo integral

Resumo da oferta

Syone is seeking an AI Platform Engineer to join our team in Lisbon. You will design and operate an enterprise AI platform, integrating LLM providers like Azure, Claude, and OpenAI, with robust routing, cost controls, and governance.

You will develop autonomous agents and multi-agent workflows, connect AI copilots to business systems, and build RAG and knowledge retrieval capabilities for grounded outputs. Strong production-focused experience is essential.

Qualificações

• 8+ years in software engineering and shipping production AI applications.
• Proven experience deploying enterprise LLM solutions used by end-users at scale.
• Experience designing multi-agent systems and agentic orchestration architectures.
• Experience integrating AI into business workflows, not just demos.

Responsabilidades

• Design and operate the enterprise AI platform layer with multi-provider LLM integrations across Azure AI Foundry, Azure OpenAI, Claude, and OpenAI APIs.
• Develop autonomous agents and multi-agent workflows that connect to enterprise platforms (SharePoint, Salesforce, M365).
• Build retrieval and knowledge-grounding capabilities with hybrid search and CPC integration.
• Ensure safety, governance, monitoring, and production readiness across AI systems.

Conhecimentos

Software engineering

Production AI apps

Enterprise LLM deployment

Multi-agent systems

Business workflows integration

Ferramentas

LangGraph

Semantic Kernel

LangChain

LlamaIndex

AutoGen

CrewAI

Datadog

Grafana

Azure AI Search

RLHF / fine-tuning

Descrição da oferta de emprego

We are looking for an AI Plarform Engineer to join our team.

Responsibilities:

• 1 — AI Platform & LLM Integration Design and operate the enterprise AI platform layer, integrating and managing multiple LLM providers in a production environment:
  • Multi-provider LLM integration across Azure AI Foundry, Azure OpenAI, Claude (AWS Bedrock), and OpenAI APIs — with routing, fallback, and cost controls
  • Reusable AI services and APIs consumed by agents, copilots, and business applications
  • Context management, memory architectures, structured outputs, and function calling
  • Inference performance optimisation, cost attribution, and usage analytics
  • AI environment management — prompt versioning, model versioning, deployment pipelines
• 2 — Agent & Agentic Workflow Development Design and deploy autonomous agents and multi-agent systems that execute real business workflows:
  • Autonomous and planning agents capable of multi-step reasoning over enterprise knowledge
  • Multi-agent orchestration using Model Context Protocol (MCP) and agent collaboration frameworks
  • Tool-using agents with function calling, plugin architecture, and enterprise system integrations
  • AI copilots built on Microsoft Copilot Studio for business users
  • Workflow automation connecting AI agents to enterprise platforms (SharePoint, Salesforce, M365, etc)
• 3 — RAG & Knowledge Retrieval Build enterprise-grade retrieval capabilities on top of the Corporate Brain data platform:
  • Hybrid search (semantic + keyword) with vector indexing and Azure AI Search integration
  • Context retrieval, citation frameworks, and prompt orchestration for grounded, accurate outputs
  • Chunking strategies, embedding pipeline consumption, and knowledge grounding
  • RAG evaluation — precision, recall, hallucination detection, and context relevance testing.
• 4 — Evaluation, Safety & AI Governance Own the quality and safety bar for all AI systems in production — this is a first-class engineering responsibility, not an afterthought:
  • Systematic evaluation frameworks covering accuracy, hallucination detection, and response quality
  • Prompt regression testing, agent benchmarking, and safety testing across model updates
  • Guardrails design and implementation — input/output filtering, content policy enforcement
  • Production monitoring: latency, cost per query, error rates, and agent task completion rates
  • AI lifecycle governance — structured approval gates before production deployment.

Required Technical Skills:

Experience:

• 8+ years in software engineering, with at least 3 years building and shipping production AI applications
• Proven experience deploying enterprise LLM solutions used by real end-users at scale
• Experience designing multi-agent systems and agentic orchestration architectures
• Experience integrating AI into business workflows — not just standalone demos.

Nice to Have:

• LangGraph
• Semantic Kernel
• LangChain or LlamaIndex
• AutoGen / CrewAI
• Observability platforms (Datadog, Grafana)
• AI governance frameworks
• Azure AI Search (advanced)
• Fine-tuning or RLHF experience

What can Syone offer me:

Integration in an organization with profound and sustained growth and involvement in pioneering projects with innovative technological solutions;

Strong IT training plans;

Professional evolution with intervention in ambitious technological projects, both national and internationally.

By applying, you accept our GDPR policies. Your personal data in your CV and documents will be used solely for processing your application at SYONE. SYONE, S.A. (NIPC 504729624) is the data controller, located at Rua Alfredo da Silva, 8-A, Edifício Stern, Piso 3D, 2610-016 Amadora. We will use your data to tailor your application to relevant vacancies and for potential employment contracts. If selected, your data will be kept during your employment and for at least two years after. If not selected, your data will be kept for two years for future vacancies and then deleted, except for your contact details. You can exercise your rights (access, rectification, erasure, etc.) by emailing [e-mail]. Your data may be shared with our HR service providers and will be stored securely. You can withdraw your consent at any time. We will not use your data for other purposes without your express consent. For more information, email [e-mail].