AI Agent Platform Engineer
KUBO International AB · Göteborg · Tills vidare · Deadline 2026-12-20
KUBO International AB is engaging an **AI Agent Platform Engineer** in Gothenburg for an engineering project within the aviation industry. Working to transition AI from prototypes into dependable production use, the role involves building and operating the underlying platform for AI agents. Key duties include contributing to platform design, configuring secure delegated access, maintaining multi-tenant infrastructure that enforces regional data residency, establishing secure multi-account AWS deployment pipelines, and introducing safeguards and tracing to manage model costs and agent behaviour. The position offers a **flexible hybrid arrangement**, typically requiring two to three days per week in the office, with the option of regular employment or contracting through one's own company. The ideal candidate brings hands-on experience running LLM applications or AI agents in production, alongside **strong AWS experience** covering infrastructure as code, private networking, and multi-customer environments. Candidates must also have practical expertise in delegated access and token flows, multi-account CI/CD pipelines, and tracing and monitoring production systems across distributed services. Strong collaborative skills to support engineering teams without direct line management are essential, while experience with Auth0 is considered a plus.
Ready to help AI take flight — and keep it on course? Join an engineering project in the aviation industry, where the challenge is taking AI beyond the demo and into dependable, everyday use.
As an AI Agent Platform Engineer, you’ll build the foundations that help teams bring agents to production safely, understand how they behave and keep their costs under control. You’ll combine hands-on engineering with a voice in platform design, working alongside experienced engineers and the teams building AI applications. Apply today and help turn AI ambition into production reality!
Key responsibilities
- Build and run the platform that supports AI agents in production.
- Contribute to platform design decisions and help turn them into working solutions.
- Set up secure access so agents can act on behalf of users with only the permissions they need.
- Build secure deployment pipelines across AWS accounts and regions, including signing and scanning shared software components.
- Maintain infrastructure that keeps customers’ environments separate and their data in the required regions.
- Introduce monitoring, tracing and safeguards to track agent behaviour, limit repeated actions and control model costs.
- Support teams building AI agents, helping them use the platform and resolve technical challenges.
Ideal candidate profile
- Hands-on experience running LLM applications or AI agents in production, beyond prototypes and demos.
- Strong AWS experience, including platforms serving multiple customers, private networking and infrastructure as code.
- Practical knowledge of delegated access and on-behalf-of token flows, so agents can act within a user’s permissions.
- Experience building secure CI/CD pipelines, including deployments across multiple AWS accounts and regions.
- Experience monitoring production systems, investigating issues and tracing activity across services.
- A collaborative, hands-on approach and the ability to help different engineering teams move forward without directly managing them.
- Nice to have: experience with Auth0.
Conditions
- Location: Gothenburg, Sweden,
- Work model: Flexible hybrid arrangement (typically, 2-3 days from office)
- Contract: Regular employment or through your own company (your choice)
Recruitment steps
- Initial call with a KUBO recruiter to discuss your experience, strengths and development interests.
- Two interviews with the client.
- Decision and feedback.