About Combient
Combient Group brings together leading large companies in the Nordics on the digitalization journey. We are convinced that collaborating across industries helps us move smarter and faster. Our network collectively includes 36 large enterprises, 1.4 million employees, and €270 billion in revenue.
As AI technologies proliferate rapidly, long-standing business operating assumptions are being fundamentally reshaped. We are the first generation of leaders navigating this transition in real time. In response, AI @ Combient brings together initiatives and services that help companies move from uncertainty to capability—through learning, strategic advisory, and practical interpretation of fast-evolving technology. Within this ecosystem, the Combient AI Center is a newly established applied research arm that combines deep technical expertise with the informed judgment that comes from hands-on building.
Our 2026/2027 roadmap involves three streams of work in collaboration with our network to generate both evidence and product for our companies, including
- controlled experimental study focused on engineering-agent collaboration,
- fast prototyping cycles to evaluate emerging ideas in generative and agentic systems, and
- AI product engineering to address challenges we have surfaced from the network of companies.
We focus on the collective priorities and challenges of the network by building real systems, testing them in live organizations, and scaling successful pilots into sustained operating capabilities.
The Role
We’re hiring an AI Engineer and an Software Engineer to be founding members of the Combient AI Center in Stockholm.
This role is for someone who enjoys working where research meets production: designing experiments, building prototypes that generate evidence, and turning the strongest concepts into results into production outcomes, reusable products, and evidence that teams and leaders can rely on.
You’ll work at the intersection of:
- LLM systems (RAG, tool calls, agents) and evaluation
- Data + evaluation infrastructure
- Human-in-the-loop system design
- Enterprise constraints (privacy, governance, security, incentives)
The work is hands-on, collaborative, and social by design. You’ll build things with people from multiple companies, always with shared ownership of the outcome.
What You’ll Do
You will contribute to and/or lead work such as:
- Design and run applied studies of AI systems in real workflows (define hypotheses, baselines, success metrics, and instrumentation).
- Build evaluation and telemetry infrastructure to measure outcomes like productivity, rework, quality, maintainability, and trust—without relying on vanity metrics.
- Develop agent-assisted workflows and prototypes that can be tested under constraints (reviews, CI/CD, security boundaries, approval processes).
- Create reference implementations and “how-to” patterns that associated companies can adopt with confidence.
- Collaborate closely with researchers and social science/UX partners to connect technical signals with human behavior and organizational realities.
- Communicate findings through internal writeups, demos, and practitioner sessions
What We’re Looking For
You don’t need to tick every box, but we look for candidates who have a growth mindset and can adapt to the moment. Most strong candidates will have:
- Strong software engineering fundamentals (Python + production systems).
- Hands-on experience with shipping LLM-backed products, retrieval, monitoring.
- Comfort with experimental thinking: measurement design, baselines, confounders, and iteration.
- Experience building robust data pipelines / logging / observability for AI products.
- Ability to operate in ambiguity and collaboratively define the right problem.
You enjoy collaborating, sharing, and building things others can actually use.
Bonus Points
- Experience in developer tooling, CI/CD, or workflow instrumentation
- Experience in regulated environments or privacy-preserving data practices
- Publications are welcome but not required (we care that you can produce credible evidence, not just PDFs)
How We Work
- We value respectful debate and evidence-based decisions
- We compare and contrast approaches, make trade-offs explicit, and time-box discussion
- Once we decide, we align and execute with shared ownership
Why This Role Is Different
- Your work won’t end at a demo: you’ll generate evidence and turn it into reusable systems.
- You’ll collaborate across multiple large organizations, so your impact compounds.
- You’ll help shape what “responsible, measurable AI adoption” looks like in practice.
Practicalities
- Hybrid work arrangement, based in Stockholm, Sweden
- Occasional travel to associated companies in Nordics
- Strong peer group across engineering, product, and research/social science
- High autonomy and influence over technical direction