About the Position

Join a multinational company with a startup spirit. You can make the difference!

At Keyrus, we help organisations move from experimental AI to industrialised AI, from isolated agents to orchestrated systems, and from insight to execution. This is the discipline we call being an Architect of Intelligence. Designing the Operating System of the intelligent enterprise, where intelligence is embedded into the core of business processes to create sustainable value: we operationalise intelligence.

AI does not transform businesses. Architected intelligence does.

Responsibilities

As an AI Engineer, you will help organisations transform complexity into measurable outcomes by combining technology, data, intelligence, and human decision-making.

Design and build production AI capabilities, including RAG systems, intelligent assistants and agent-enabled workflows aligned with the approved solution architecture.
Engineer agent behaviour across instructions, context, tools, state, routing, memory, recovery, stopping conditions and appropriate human checkpoints.
Integrate and configure models through APIs or supported platforms, making the right trade-offs between capability, quality, latency, cost and operational requirements.
Design evaluations and quality gates using representative datasets, automated and regression testing, task-success measures, groundedness and other relevant indicators.
Deploy, monitor and improve AI capabilities in production, using evaluation results and observed system behaviour to drive continuous improvement.
Implement reliability and control mechanisms such as validation, guardrails, failure handling and human oversight, within the architecture and governance standards defined for the engagement.
Integrate AI capabilities with the surrounding solution and client systems, working with Software Engineers on application interfaces and with Data Engineers on retrieval, data quality and AI-ready data.
Enable client teams to run and extend the agents after go-live.
Document and reuse what works, including AI behaviour, dependencies, evaluation criteria, limitations and engineering patterns that strengthen future Keyrus engagements.

Requirements

Degree in Computer Science, Engineering or a related field, or equivalent professional experience.
Strong hands-on experience building production GenAI / LLM-based systems, beyond notebook-based or demonstration work.
Experience deploying and operating AI capabilities in production, including monitoring, reliability, latency, cost and continuous improvement.
Ability to work effectively in multidisciplinary teams alongside software, data, architecture and governance specialists.
Availability for monthly travel to Lisbon or Madrid.
Professional fluency in Portuguese, Spanish and English, written and spoken.

Technical & Professional Skills

Strong Python and software-engineering fundamentals, including APIs, testing, version control, code review and maintainable production code.
Practical experience designing and improving RAG and retrieval-based systems, including context preparation, retrieval quality, grounding and relevant data dependencies.
Hands-on experience engineering AI agents or multi-step AI workflows, including tool use, context management, routing, state, failure handling and human control.
Experience creating evaluations for LLM or agent systems, using measurable criteria and representative test cases to assess behaviour and detect regressions.
Experience integrating commercial or open-source model providers, and integrating AI with enterprise applications, APIs, data sources and workflow systems (e.g. CRM, ERP, MCP, webhooks)
Ability to tie agent behaviour to the business KPI it serves.
Working knowledge of solution-level guardrails, validation, human oversight and responsible AI engineering practices.
Familiarity with at least one major cloud or AI ecosystem, plus containerisation, CI/CD, observability and LLMOps practices, and the ability to optimise for quality, latency, context usage and inference cost.
Depth in one or more ecosystems such as LangGraph, LangChain, LlamaIndex or Semantic Kernel; OpenAI, Anthropic, Google, Microsoft or open-source models; and pgvector, Pinecone, Weaviate or Milvus. These are examples, not a checklist.

Nice to Have

Deeper agentic AI experience, including multi-agent orchestration, planning and execution loops, complex state management, recovery strategies and sophisticated human-in-the-loop patterns.
Experience with multimodal systems, fine-tuning, open-source model deployment or specialised inference.
Classical machine-learning or deep-learning experience.
Low-code agent platforms such as n8n, Make or Dust for fast automations alongside code.
Experience in regulated or security-sensitive environments.
Consulting or client-facing experience, and exposure to international projects or multicultural teams.
Experience mentoring engineers or creating reusable AI engineering standards, components or evaluation approaches.

We Offer

Competitive salary aligned with your experience and the data market
Meal allowance: €10.20/day
Flexible benefits plan
Private medical insurance
22 days of annual leave, increasing every 3 years (up to 25 days)
Continuous learning via KLX – Keyrus Learning Experience
A collaborative, international, and human-centred work environment

About the Company

AI does not transform businesses. Architected intelligence does.

Keyrus Portugal participates in the +Talento Internship Program, co-funded by IEFP and Portugal 2030*.*

AI Engineer

Job description