CS MSc Thesis Presentation 1 June 2026

Monday, 1 June there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:2405 (Glasburen).

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname [dot] lastname [at] cs [dot] lth [dot] se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions for opponents are found here on the LTH thesis project page.

11:15-12:00 in E:2405 (Glasburen)

• Presenters: Biyu Zou, Zhe Zhang
• Title: AI and Automation Driven Workflow Optimization for Solution Packages
• Examiner: Görel Hedin
• Supervisors: Pierre Nugues (LTH), Anne Sjögren, Lifeng Han (Ericsson AB)

In Ericsson’s Solution Package (SP) workflow, critical project data is scattered across systems like Confluence, Jira, and FPT. This severe information fragmentation forces coordinators (SP Drivers) to manually cross-check statuses, risks, and evidence.

This thesis presents SP Copilot, an AI-assisted prototype tailored to streamline project coordination. Addressing enterprise requirements for traceability and low latency, the system utilizes a hybrid architecture integrating deterministic rule-based heuristics, exact glossary indexing, and constrained LLM synthesis, explicitly departing from conventional RAG. Furthermore, it leverages the Model Context Protocol (MCP) to cleanly decouple system integrations.

The copilot demonstrates five core skills: multi-source summarization, human-in-the-loop synchronization, diagnostic risk assessment, task prioritization, and offline knowledge lookup. Ultimately, this work provides an architectural blueprint for enterprise AI copilots that reduce manual overhead, ensure evidence grounding, and enhance decision-making without compromising operational governance.