Master Thesis – Establishing Trust in AI Through a Practical Framework

ASSA ABLOY Opening Solutions

. Design and validate a practical framework for evaluating the trustworthiness of AI-powered solutions .

Posted 10/19/2026internshipStockholm • SwedenMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in evaluating AI-powered solutions with a focus on data trustworthiness, answer trustworthiness, and system reliability. Proficient in designing evaluation frameworks, benchmark datasets, and metrics for assessing AI performance and governance.

Highest-signal resume keywords

AI-Powered Solutions EvaluationData Trustworthiness AssessmentBenchmark Dataset DesignTrustworthiness Metrics DefinitionRegression Testing and Monitoring

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Evaluation Framework DesignAutomated Test Harness DevelopmentRisk Threshold DefinitionAdversarial Prompt DesignSemantic Similarity TechniquesFactual Accuracy MeasurementConfidence CalibrationLatency MeasurementCost AnalysisFeedback Collection

Soft Skills

CollaborationCritical ThinkingProblem SolvingCommunication

Tools & Technologies

Graph RAGKnowledge EngineeringOntologiesKnowledge Orchestrators

Industry Keywords

AI GovernanceUser Trust ExperienceModel Behavior EvaluationData Quality BenchmarkingContinuous Evaluation

Tech Stack

Tools & technologies

Spring

About the role

Key responsibilities & impact

Design and validate a practical framework for evaluating the trustworthiness of AI-powered solutions
Create an automated evaluation mechanism with measurable benchmarks for data quality, model behavior, and answer quality
Develop a representative evaluation dataset containing relevant questions, expected answers or acceptance criteria, known edge cases, and deliberately misleading inputs
Evaluate AI solutions repeatedly as data, prompts, models, or system design change
Assess data trustworthiness, answer trustworthiness, system reliability, safety and governance, and user trust experience
Identify key stakeholders needed to build trust in an AI solution
Verify output relevance using rule-based checks, retrieval-source validation, domain-expert review, LLM-as-a-judge techniques, semantic similarity, and trusted reference answers
Detect and help prevent hallucinations, unsupported claims, fabricated citations, incorrect summaries, contradictions, and unsupported answers
Define risk thresholds for human approval, review, correction, or escalation
Design and maintain representative benchmark datasets, including adversarial prompts and real user scenarios
Define trustworthiness metrics such as groundedness, factual accuracy, answer relevance, citation correctness, consistency, completeness, refusal quality, confidence calibration, fairness, latency, and cost
Support continuous post-release evaluation through regression testing, production monitoring, feedback collection, drift detection, periodic benchmark runs, and alerting
Produce a practical evaluation framework, benchmark design method, automated test harness, scoring model, and governance recommendations
Explore advanced AI capabilities with the team, including agentic AI, Graph RAG, knowledge engineering, ontologies, and knowledge orchestrators

Requirements

What you’ll need

Master’s student
Must be available to collaborate during the spring semester of 2027
Submit a resumé and application describing ideas for approaching the thesis subject
Most work will be done at one of ASSA ABLOY’s two sites in Stockholm
Ability or interest in evaluating AI-powered solutions across data trustworthiness, answer trustworthiness, system reliability, safety and governance, and user trust experience
Interest or knowledge in AI-powered solutions, agentic AI approaches, Graph RAG applications, knowledge engineering, ontologies, and knowledge orchestrators

Benefits

Comp & perks

Cutting edge technology
A steep learning curve
Dedicated supervisors from within the enterprise
A group of fellow master's thesis students to share the experience with
Training and development opportunities
Regular feedback
Diverse and inclusive teams