AI Engineer

Uniplay; Stockholms kommun; On-site; SEK 900,000 - 1,300,000; Full time

Job summary

Uniplay, Stockholm-based, is seeking an AI Engineer to build and ship adaptive learning AI systems. You will shape agent workflows, grounding, RAG, and learning science to turn company material into game-based learning experiences. This hands-on role values rapid prototyping and production delivery.

You will collaborate with engineers, product, design, and learning experts to push the product forward and influence the roadmap with practical technical input.

Qualifications

• Experience shipping AI systems in production with real users.
• Depth in modern AI stack: RAG, agentic workflows, structured generation.
• Proven ability to turn ideas into working prototypes and production.
• Strong coding in TypeScript and/or Python; readable, maintainable code.
• Excellent written and spoken English; Swedish a plus.

Responsibilities

• Build AI learning pipelines end to end from source to generated content.
• Work with messy real-world knowledge sources to create useful output.
• Develop evaluation infrastructure to measure content quality.
• Improve AI production system: prompts, templates, orchestration, costs.
• Influence roadmap with concrete technical input and tradeoffs.
• Collaborate with engineers, product, design, and learning experts.

Skills

LLM production

RAG workflows

Knowledge graphs

TypeScript/Python

Prototype to production

English communication

Job description

At Uniplay, our vision is simple: Make Corporate Learning Fun.

We're building something different — an AI-native company where games, data, and hyper-personalisation redefine how people learn and grow at work.

Our product blends games, behavioral data, and AI agents into a living learning experience that adapts to employees and evolves with every organisation.

We care less about titles and more about people who create, collaborate, and push limits.

The role

AI isn't a feature we bolted on — it's the product. Agent workflows, source grounding, RAG, structured generation, and learning science are how we turn real company material into adaptive, game-based learning.

We're looking for an AI Engineer to join the team building that machinery: researching what's possible, building it, evaluating it honestly, and shipping it to customers. It's a hands-on role for someone who moves from paper to prototype to production without needing three planning meetings first.

What you'll do

• Build AI learning pipelines end to end. From ingestion of source material to generated, adaptive learning content — research, build, ship, measure, improve, together with the rest of the engineering team.
• Work with messy, real knowledge sources. Documents, links, videos, internal company material. Turning that into something useful is the hard part, and it's what the team spends its days on.
• Build our evaluation infrastructure. Quality in generated learning content is not a vibe. Help us define how we measure it and make the numbers move.
• Improve the AI production system. Prompts, templates, orchestration, model selection, latency, cost, internal tooling. Find the bottlenecks and remove them.
• Add a strong voice to the roadmap. You'll have real influence over what we build next — and a direct line to the people making those calls.
• Work closely with the rest of the team. Engineers, product, design, and learning experts. You'll review each other's work, pair on hard problems, and share what you learn.

What we're looking for

• Real experience shipping AI systems. You've built LLM-powered products that ran in production, with real users and real failure modes — not just notebooks and demos.
• Depth in the modern stack. RAG, agentic workflows, structured generation, evals, vector databases, knowledge graphs. You know where each one earns its keep and where it doesn't.
• Your own projects. You build things because you can't help it. Show us at least one — repo, demo, write-up, product. What you learned matters more than how polished it is.
• Extreme drive. You don't wait for permission. You make things happen and bring people along.
• Fast iteration. You prototype quickly, kill your own ideas when the data says so, and are comfortable with messy early-stage exploration.
• Strong engineering. TypeScript and/or Python. You write code others can build on.
• Curiosity about how people learn. The bottleneck to AI value isn't technology — it's human learning and behaviour change. If that thesis interests you, we'll get along.
• A team player. You share context, ask for help early, give useful feedback, and would rather the team ship something good than be right on your own.
• Clear communication in English. Swedish a plus.

Why Uniplay

• Work on AI that turns company knowledge into learning experiences people voluntarily come back to.
• Global from day one — Sweden, Brazil, UK, DACH.
• A small team where your work reaches production in days, not quarters.
• Direct collaboration with engineers, AI specialists, HR experts, and product designers.
• Stockholm-base