Thesis: Agentic AI-Ops: Autonomous AI Agents for Production Infrastructure

AI agents can interact with live infrastructure, but the safe limits of their autonomy remain uncertain. This thesis explores how different degrees of autonomy affect the reliability and safety of infrastructure operations in a student-built sandbox.

The assignment

• Build an isolated test environment comprising a Kubernetes cluster and a Jira board, and connect AI agents using MCP tool calls or an alternative approach.
• Evaluate agents at different autonomy levels: monitoring and escalation, proposing actions for human approval, and independently performing narrowly defined, low-risk operations.
• Introduce controlled incidents such as pod crashes, resource shortages and network failures, then compare agent diagnoses, actions and tickets with a human baseline.
• Investigate objective measures of root-cause diagnosis accuracy and assess the usefulness and correctness of automatically created Jira tickets.
• Explore the architecture, safeguards and permissions, including Kubernetes RBAC, required for safe integration into DevOps workflows.
• Focus on one research question at bachelor's level or conduct a broader comparative evaluation at master's level.

Who we are looking for

• Bachelor's or master's student in computer science or engineering, expecting to graduate in June 2027.
• Fluent in both Swedish and English, in speech and writing.

Practical details

Based at Knightec Group's Sundsvall office. Start in January or March 2027. Apply through the company's careers page as soon as possible, by 30 November 2026. Questions can be sent to johanna.edstrom@knightec.se.