AI Software engineer

Vessla; Stockholms kommun; On-site; SEK 900,000 - 1,400,000; Full time

Job summary

Vessla is building the next generation of AI-enabled operations for property teams, enabling agents to act reliably across connected software. You’ll help define how AI agents are used in an industry that shapes everyday life, with a direct hand in what we build and how customers experience it.

You’ll take meaningful ownership from the start, focusing on agent infrastructure, data models, evaluation systems, and customer discovery, while collaborating with product and design to create intuitive

Responsibilities

• Define how AI agents get used in property operations and ensure it works reliably
• Own meaningful ownership from the start and shape your focus around strengths and interests
• Build agent infrastructure, orchestration, and tools for complex, reliable work
• Develop data models and integrations that connect essential software and information
• Shape an intuitive interface for users and provide visibility and control

Job description

We’re building a future where property teams spend less time managing software and more time creating better places to live and work. Vessla is the next generation of agentic OS for property operations, connecting to the systems property teams already use to enable bleeding edge AI capabilities across all of them.

Vessla’s agents find the information that matters, apply each organization’s operating rules and carry out work through connected software. The hard part is making this work reliably. That means deep integrations, a detailed understanding of each business, rigorous evaluation and clear boundaries on what agents can do. Our founding team brings experience building and deploying advanced AI systems at QuantumBlack, McKinsey, NASA, Andon Labs and Legora, to tackle that challenge.

The role

Join our early team and help define how AI agents get used in an industry that shapes everyday life. You’ll have a direct hand in what we build, how it works and how customers experience it, with the scope to take an idea from a user conversation all the way into production.

You will get a chance to work on:

• Agent infrastructure: Build the orchestration and tools that let agents carry out complex work reliably
• Integrations and data models: Connect the software and information that building operations depend on
• Agentic evaluation systems: Develop ways to test agent behavior, understand failures and improve performance
• Customer discovery: Work directly with end-users to uncover problems worth solving and test solutions
• Product and design: Shape an interface that makes powerful capabilities intuitive and gives users visibility and control

You’ll take meaningful ownership from the start, with room to shape your focus around your strengths, interests and the problems where you can make the greatest difference. Our stack includes Python, TypeScript, Langfuse, AWS and TanStack.

Who we’re looking for

• You’ve built LLM systems and understand where, and why, they fail
• You enjoy talking to users and turning their problems into thoughtful product decisions
• You care about building maintainable software while moving quickly
• You bring ideas, challenge assumptions and take the initiative to try things out

We keep the path from idea to working product short. You’ll have the freedom to experiment, the responsibility to make things work, and a front seat in building the company that will define the next era of property operations.