Master Thesis - Semantic Guardrails - Autonomous Networks - Verification of Constraint Preservation

• Join our Team

About this opportunity:

Future autonomous networks will increasingly rely on AI agents that transform high-level operator intents into network optimization decisions and executable configurations. We need to ensure that during all the transformations the associated guardrails and constraints are verified and enforced. In this master thesis you have the opportunity to explore the creation and use of semantic equivalence chains (e.g. "I want to close my account" -> "user-terminated subscription" -> POST /api/v1/billing/cancel) to verify that guardrails are followed and constraints are not broken.

What you will do:

You will need to create an AI-based "Semantic Assurance Agent" that will check that any decisions (i.e. agentic transformations) based an initial intent respect the original constraints and guardrails. You will explore creating and using semantic equivalence chains (starting from the decision input or output) to attempt to recognize additional guardrail deviations and constraint breaches than a direct input and output comparison. You will also need to explore the definition of what an "acceptable" guardrail deviation is.

The skills you bring:

• General Programming skills
• An inquisitive mindset
• Knowledge of AI concepts or an active interest to independently expand your knowledge on the relevant topics.
• Designing the experimental setup and the baseline comparison of your "Semantic Assurance Agent" output.

Why join Ericsson?At Ericsson, you´ll have an outstanding opportunity. The chance to use your skills and imagination to push the boundaries of what´s possible. To build solutions never seen before to some of the world's toughest problems. You´ll be challenged, but you won't be alone. You´ll be joining a team of diverse innovators, all driven to go beyond the status quo to craft what comes next.

Primary country and city: Sweden (SE) || Stockholm

Req ID: 791334