Master Thesis in Software Engineering: Adaptive Architecture for Human-AI Annotation Systems

Technology, Data & Digital · Software & Web Development · Software Engineering

In short

Master's students sought for a thesis on designing and implementing a confidence-aware adaptive verification architecture for Human-AI annotation systems. The project focuses on software engineering, evaluating an existing platform, and understanding human-AI interaction in autonomous transport solutions. Preferred start date January 2027.

Responsibilities

Design and integrate a modular architecture with explicit interfaces for confidence providers, verification policies, decision logging, and annotator feedback.
Implement confidence-provider components exposing calibrated or normalized signals (e.g., model confidence, model agreement, source model, object class frequency).
Implement and compare interchangeable threshold-based and learned decision policies for routing cases to light review, full review, or a second review.
Integrate attention cues into the annotation interface and record relevant signals and outcomes.
Develop automated tests for routing decisions, component interfaces, and fallback behavior.
Benchmark the adaptive architecture against existing workflows using metrics like annotation correctness, recall on safety-critical classes, error catch rates, and task completion time.
Evaluate maintainability and testability through change scenarios.
Conduct a user study to investigate verification behavior, cognitive workload, trust calibration, and susceptibility to automation bias.
Explore out-of-distribution detection or vision-language-model-derived scene complexity as additional confidence signals (optional).
Consider online adaptation as an optional extension.

Requirements

Final-year Master's student in Software Engineering, Computer Science, or a related field.
Strong interest in software architecture, software quality, and building maintainable systems.
Comfortable programming in Python.
Foundation in modular software design.
Foundation in interface design.
Foundation in automated testing.
Foundation in version control.
Interest in empirical software engineering.
Motivation to compare design alternatives and connect architectural decisions with measurable effects.
Interest in learning about model confidence and human verification to make sound software design decisions.

Desired Qualifications

Familiarity with AI-assisted systems.
Familiarity with machine learning.
Familiarity with computer vision.
Familiarity with human-computer interaction.
Prior specialization in machine learning is not required.
Prior specialization in object detection is not required.

Benefits

Opportunity to apply software engineering methods to a concrete industrial challenge.
Design an adaptive human-AI workflow that is maintainable, testable, and auditable.
Build on an existing annotation platform and prior thesis results.
Gain hands-on experience in architectural design, component integration, automated testing, and evaluation of an industrial software system.
Investigate how confidence signals can guide human attention while preserving traceability and human oversight.
Measure architectural trade-offs across annotation correctness, performance, maintainability, and testability.
Collaborate with engineers working on the validation of autonomous transport solutions.
Gain practical experience in designing and evaluating human-AI systems in a safety-critical context.

Skills

PythonSoftware ArchitectureAutomated TestingVersion ControlHuman-Computer InteractionMachine LearningComputer Vision

#Software Engineering#Human-AI#Annotation#Architecture#Autonomous Systems#Volvo#Master Thesis

About Volvo Group

The Volvo Group is one of the world’s leading manufacturers of trucks, buses, construction equipment and marine and industrial engines. The Group also provides complete solutions for financing and service. The Volvo Group, with its headquarters in Gothenburg, employs about 100,000 people, has production facilities in 18 countries and sells its products in more than 190 markets.\n\n

See all open jobs at Volvo Group Read more about Volvo Group Website