AI Solutions Architect

Curamando, part of the Nordic consulting house Eidra, specialises in accelerating top-line growth through digital strategy and innovation. We're a fast-growing consultancy already partnering with industry leaders - from cutting-edge automotive brands to forward-looking tech companies. Together with Eidra's 1,200+ strategists, creatives, and engineers across Stockholm, New York, and beyond, we tackle ambitious problems at scale.

Our AI practice is growing fast, and we are now looking for the person who will be the architect behind our next generation of AI builds.

Job Summary

We are looking for a hands-on AI Solutions Architect with a proven track record of designing and delivering AI solutions that make it into production. You have been the architect behind several AI projects already, and you can walk into a client situation and say - with the calm of someone who has done it many times - how this should be built: what the solution looks like, what team it needs, how long it will take, and in what stages it should be rolled out.

This is a senior, technical, client-facing role. It is not a people-leadership role, and it is not a project-management role. You will shape the scope and the architecture of the whole project, and then roll up your sleeves and be part of building it.

What you will do

- Architect the build. Design AI solutions that fit the client's problem, data and organisation - increasingly agentic systems with context, memory and tool handling on top of a solid data foundation. Understand the trade-offs, communicate them clearly, and help the team choose the right technology.
- Scope and size projects. Turn a client's ambition into a concrete plan: the solution architecture, the team composition, the timeline, the phases and rollout stages, what is in and out of scope, and how to phase and price it. This is the core of the role.
- Be part of building. Not just the diagram - the build. Go into projects, work alongside our AI engineers, review the critical pieces, unblock the team, and make sure the architecture survives contact with reality.
- Be the client's technical counterpart. Establish trust as a recognised AI authority, guide clients through the possibilities, limitations, suitability and risks of AI solutions, and enjoy talking about the value you create. You will contribute to proposals and pitches alongside our commercial leads.
- Own the journey to production. Break complex challenges into manageable parts, prioritise the key deliverables, identify and resolve bottlenecks, catch the overlooked details, and anticipate what could go wrong before it does.
- Design for Operate and Transfer. Build with operability and hand-over in mind from day one, and stay engaged as the solution moves from build into operation and, eventually, into the client's own hands.
- Stay ahead of the curve. Follow the development of the AI stack and the regulatory landscape across the markets we work in, and bring that knowledge into architectural decisions.

Who you are

- You have a solid technical background , most likely from the software or tech sector - a product company, a scale-up, or a technology consultancy - and you have been the architect behind several AI projects that reached production.
- You've done the hands-on work. Whatever your current title - engineer, architect, tech lead, technical product manager - you have built things hands-on at some point in your career, and you can still go deep when it matters.
- You can size a project. Given a problem, you can quickly and credibly say what team is needed, how long it will take and how it should be staged - because you have done it many times before.
- You have a strong intuition for AI , and you know how to balance its transformative potential against its practical limitations.
- You are familiar in the modern AI stack : You understand the purpose and place where LLM-based and agentic architectures fit (frameworks such as LangChain/LangGraph, Mastra or Microsoft Agent Framework), evaluation and observability tooling (Langfuse, Promptfoo or similar), data pipelines and MLOps - and you know the classical ML side (PyTorch/TensorFlow, fine-tuning, PEFT) well enough to know when it is the right tool.
- You are commercially minded and enjoy talking to clients - about their problem, and about the value you bring. You do not need to be a people leader, but this is not a role for someone who would rather sit behind a closed door with a computer.
- Consultancy experience is a plus , not a must.