Recent and upcoming events

DATE: October 30th, 9-10:30 am
We are excited to have <a href="https://people.csiro.au/L/Q/Qinghua-Lu">Qinghua Lu</a> as a virtual speaker for the AI Engineering Lab.&nbsp; Qinghua will join us remotely from Australia, but we have booked 473 for an in person viewing/workshop. &nbsp;We have booked 1.5 hours, and propose a hybrid seminar/workshop, as Qinghua speaks on a topic (Agentic AI Engineering) that is very interesting for many members. Agenda, October 30th, 2025, zoom/473 9:00 Brief Intro ~40 minutes:&nbsp; Qinghua’s talk ~20 minutes: &nbsp;Talk questions/discussion
~30 minutes:&nbsp; Agentic AI Engineering Workshop – Topic: What are the future trends in Agentic AI Engineering?&nbsp; Challenges?&nbsp; Opportunities?
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ~15 minutes: break out discussions
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; ~15 minutes consolidation
<h4>Talk title: Responsible and Safe Agentic AI Engineering</h4>
Talk abstract:&nbsp;
Agentic AI engineering is an emerging discipline focused on the design, development, and operation of AI systems that exhibit goal-directed autonomy. Foundation models (FMs), such as large language models (LLMs), have been accelerating progress in this area across academia and industry, but they also amplify challenges of responsible AI and&nbsp;AI safety.&nbsp;Agentic AI systems often involve multiple interacting agents, humans, and tools, requiring rigorous system-level engineering to ensure critical qualities like robustness, safety, and observability. A key design challenge in agentic engineering is the growing capability of FMs/LLMs. Developers must decide whether to rely on the FM/LLM or external tools/systems for the same functionality. These decisions can be made at various stages depending on the problem and context: during design time, development time, or even at runtime from a software engineering perspective, and at pre-training time, post-training time, test/inference time, and post-inference time from an AI perspective. Highly autonomous agentic systems also require continuous monitoring, evaluation, observability, intervention, and oversight after deployment, an emerging discipline referred to as AgentOps. This talk will present our latest research in responsible agentic AI engineering, including pattern-oriented&nbsp;agent architecture, Swiss Cheese Model for agentic AI safety, AgentOps, and design for meaningful human oversight.&nbsp;
&nbsp;
Bio:
Dr Qinghua Lu is the group leader of the Software/AI Systems Research Group, which consists of over 40 researchers/engineers across 5 research teams. She is widely recognised as one of Australia’s leading scientific experts in Responsible AI and AI Engineering and plays a key role in representing the Australian research community in major national and international AI safety initiatives including Australia’s AI Safety Standard, International Network of AI Safety Institutes, Frontier Model Forum (together with OpenAI, Microsoft, Google, Meta, Anthropic), OECD.AI, and EU General-Purpose AI Code of Practice. She has also provided technical expert input to support Australia’s representation in international forums, including contributing to the Bengio-led International AI Safety Scientific Report, supporting the Australian representative.
Dr. Lu has published over 200 papers in premier international journals/conferences. Her citation count exceeds 10k+ and her h-index is 51. She is an editor for several top AI/SE journals, including IEEE Transactions on AI. She is also the organisation/program committee member of top AI/software engineering conferences, including IJCAI, AAAI, NeurIPS, ICML, ICSE, FSE, CAIN, AIware etc.&nbsp;
Dr. Lu was the winner of the 2023 Asia-Pacific Women in AI Trailblazer Award. Her recent paper titled “Towards a Roadmap on Software Engineering for Responsible AI” won the ACM Distinguished Paper Award. Her seminal paper on LLM agent architecture, awarded the 2024 Best Paper Award by IEEE Software.&nbsp;Dr. Lu’s book, “Responsible AI: Best Practices for Creating Trustworthy AI Systems”, published by Pearson Addison-Wesley in December 2023, is the world’s first responsible AI book for practitioners and became Amazon’s No. 3 AI book bestseller. Her new book, “Engineering AI Systems: Architecture and DevOps Essentials” was published by Pearson Addison-Wesley in March 2025.

DATE: Nov. 27th, 13-15, Jupiter 473.&nbsp;&nbsp; In person preferred.&nbsp; &nbsp;&nbsp;
One-hour talk followed by a one-hour workshop on green ML &amp; software architecture.&nbsp;&nbsp;
Talk title:&nbsp; The Challenge of Designing High-Quality Green ML-Enabled Systems: A Software Architecture Perspective
Abstract: While artificial intelligence (AI) and machine learning (ML) techniques are increasingly maturing, it remains very challenging for industry to develop production-ready AI-based systems and to operate and maintain them for longer periods of time. Even though ML models may only be a small piece of these systems, they nonetheless have a profound impact on their functionality and quality attributes, which can be difficult to manage. One important quality attribute that has not received the priority it deserves is environmental sustainability. In this talk, I will briefly introduce how software engineering for AI-based systems (SE4AI) has emerged to try to tackle these problems. I will then take a software architecture perspective on these systems, and finally talk about Green AI by describing our past and ongoing efforts to synthesize and evaluate architectural tactics to improve the environmental sustainability of ML-enabled systems.
About the Presenter:&nbsp; Dr. Justus Bogner is an Assistant Professor in Software Engineering within the Software and Sustainability (S2) group at Vrije Universiteit Amsterdam, The Netherlands. During and after his studies as well as throughout his PhD, he worked as a software engineer at Hewlett Packard and later at DXC Technology for more than nine years. His main research interests are empirical software engineering, software architecture, and software quality (maintainability, evolvability, sustainability), typically applied to AI-based systems (SE4AI) and microservice-based systems. He is the program co-chair for CAIN'26 and co-organizes the SAML workshop on software architecture and machine learning.
Further Information:
- Personal website of the speaker: <a href="https://xjreb.github.io">https://xjreb.github.io</a>
- LinkedIn profile: <a href="https://www.linkedin.com/in/justus-bogner">https://www.linkedin.com/in/justus-bogner</a>

Please join us on September 11-12 for the 2025 WASP Software Engineering and Technology Cluster Workshop, conducted in collaboration with the Chalmers/GU AI Engineering Laboratory.
This will be an exciting opportunity to hear about the research being conducted by world class leaders in the software engineering field - both internationally and within Sweden - and to network, find new collaborations, and more!
When and Where
September 11 - 12, in room Svea 130 at the Chalmers Lindholmen campus.
The workshop will take place in the afternoon on the 11th, and all day on the 12th.&nbsp;
A full program will be announced later. Please feel free to join us for any part of the program.
&nbsp;
Registration
The workshop is free, but we ask that you register so that we have rough estimates of how much food to order.&nbsp;
Please note that the dinner is only open to members of the WASP cluster. However, all other events are open to anyone to attend.&nbsp;
You can register at:&nbsp;<a href="https://forms.cloud.microsoft/e/qx6s31fPdd">https://forms.cloud.microsoft/e/qx6s31fPdd</a>&nbsp;
Note that this workshop is open to anyone, so feel free to share this with both academic and industrial colleagues.
Keynote Speakers
September 11th, 13:30 - &nbsp;15:00:&nbsp;
Building and Evaluating AI-Powered Apps: Challenges, Lessons, and Future Research Directions
Chris Parnin, Principal Researcher, Microsoft Research
Abstract:&nbsp;The integration of advanced AI, particularly Large Language Models (LLMs), into software products is transforming how users interact with technology and how software is built. Moving beyond button clicks, a user can now expect to communicate with software using natural language—e.g., “I have extra vacation days to use up. Go ahead and create an OOF event for every Friday without any scheduled team meetings for every remaining vacation day”—and expect it to work. Traditional software architectures were not designed to handle such open-ended user interactions, introducing unprecedented challenges for software engineers. This talk draws on interviews with over 50 practitioners and a survey of 332 professionals to discuss pain points, challenges, and emerging solutions. It also connects these challenges to experiences building product Copilot at Microsoft, including Visual Studio and Office products. Finally, it explores the challenges of evaluating AI-powered software, including the role of graders and LLMs-as-a-judge in task evaluation. This talk offers a comprehensive overview of the challenges, research opportunities, and actionable strategies for shaping the future of robust, user-centric AI-powered systems.
Biography:&nbsp;Dr. Chris Parnin’s research spans developer productivity, cognition and learning, and automated infrastructure. He has published over 90 papers, received five SIGSOFT Distinguished Paper Awards, a 10-Year Impact Paper Award, a Google Faculty Award, and an NSF CAREER Award. His work has been featured in hundreds of international news articles, magazines, and frequently discussed in industry forums. Currently, Dr. Parnin leads research efforts at Microsoft focused on building and evaluating Copilots for Visual Studio and the Office Suite. His experience also includes serving as a tenured professor at NC State University, contributing to Microsoft Research's Human Interactions in Programming group, conducting field studies with ABB Research, and over a decade of professional programming in the defense industry. Dr. Parnin has collaborated with industry partners to bridge research and practice, including developing mentorship programs for Stack Overflow, creating anxiety-reducing features for CoderPad, and advancing equitable hiring solutions with Byteboard. He also co-organized five Continuous Deployment Summits, hosted by leading companies such as Facebook, Netflix, Google, Microsoft, and Twitter. These summits fostered collaboration and innovation in continuous deployment practices, resulting in impactful publications and educational resources for students and professionals.
September 12th, 9:00 - 10:30:
Beyond Autocomplete: Towards Effective Human-AI Interaction in Software Engineering
Christoph Treude, Associate Professor, Singapore Management University
Abstract:&nbsp;Large Language Models (LLMs) and multi-agent AI systems are rapidly becoming part of software engineering workflows, from generating code suggestions to reviewing pull requests and annotating datasets. Yet, their impact depends not just on raw model capabilities, but on how we design the interactions between humans and AI. In this talk, I will explore what it takes to get these interactions right, drawing on a series of empirical studies we have conducted over the last few years. I will discuss how developers' adoption of AI-generated code depends on what the AI chooses to present; how to design AI code review bots that lead to actual code changes rather than ignored suggestions; how to leverage LLMs to support human annotation of software engineering artifacts; and how to address the practical and ethical challenges of reusing pre-trained AI models. The success of AI in software engineering depends on integrating it into human workflows in ways informed by insights from multiple areas, including AI for Software Engineering (AI4SE), Human-Computer Interaction (HCI), and the Human and Social Aspects of Software Engineering.
Biography:&nbsp;Christoph Treude is an Associate Professor of Computer Science at Singapore Management University. His work spans empirical and automated software engineering (SE), artificial intelligence (AI) and software engineering (AI&amp;SE), human-AI collaboration, and AI for science. He has authored over 200 scientific publications in collaboration with more than 300 co-authors. His research has been recognized with five best paper awards, including three ACM SIGSOFT Distinguished Paper Awards, and has received funding from Google, Facebook, DST, and through an ARC Discovery Early Career Researcher Award (2018–2020). Before joining SMU, he held academic positions at the University of Melbourne and the University of Adelaide, and postdoctoral appointments at McGill University, the University of São Paulo, and the Federal University of Rio Grande do Norte. He currently serves on the editorial boards of the IEEE Transactions on Software Engineering and Empirical Software Engineering (Springer), and as Open Science Editor for the Journal of Systems and Software (Elsevier). He is co-chairing the program committee for FSE 2026 and regularly serves on program committees across leading software engineering venues.
&nbsp;

DATE: April 24th, 14:00 (birthday fika at 16:00)
PLACE: Lindholmen Campus, Jupiter building, room to be determined (online with Microsoft Teams).
We are excited to welcome Youri Voet, co-founder and CEO of CodeGrade.
CodeGrade's AI Assistant integrates seamlessly into your Canvas assignments, offering students guided support as they code. As an instructor you can choose from three assistant types
<ul>
<li>Learning Assistant – for conceptual help</li>
<li>Feedback Assistant – for suggestions and improvements</li>
<li>Coding Assistant – for technical coding support</li>
</ul>
Each prompt can be fully customisable and can be enabled per assignment based on your teaching goals. CodeGrade provides full visibility by monitoring student interactions to ensure ethical use and gain insights into their learning process. All AI interactions are privacy-compliant, run securely on our infrastructure via AWS and Anthropic’s Claude model.
Youri Voet's presentation will be followed by an open discussion. Don’t miss this opportunity to have a down-to-earth and solution-oriented chat on the implications of AI usage by students. Also, do not hesitate to join if you are a student yourself: your insights are valuable!
 
For more information about CodeGrade: <a href="https://www.codegrade.com/features/ai-assistant" title="https://www.codegrade.com/features/ai-assistant">https://www.codegrade.com/features/ai-assistant</a>
For more information about the AI Engineering Lab: <a href="https://www.chalmers.se/en/projects/ai-engineering-lab/" title="https://www.chalmers.se/en/projects/ai-engineering-lab/">https://www.chalmers.se/en/projects/ai-engineering-lab/</a>
Please, register for fika no later than 12:00 Tuesday April 22nd here: <a href="https://forms.office.com/e/dUaHDRa5Vf" title="https://forms.office.com/e/dUaHDRa5Vf">https://forms.office.com/e/dUaHDRa5Vf</a>

Date &amp; Time: Tuesday, 25th February 2025, 11:00 AM – 12:00 PM Location: CSE Building, Room Jupiter 473
Abstract:
While many academics are still conducting valuable research on requirements specifications, practitioners choose how to represent their requirements, depending on the context, from a variety of possible artifacts. In many cases, traditional specification documents are deemed as unsuitable. In this talk, I will introduce ongoing research regarding requirements artifacts that go beyond specifications, focusing on communication media such as oral conversations and issue tracking systems. I will argue that these informal, ad-hoc representations of requirements are an opportunity for the application of modern AI, both for researchers and practitioners, thanks to the unprecedented ability of LLMs to process unstructured and multi-modal representations of requirements.
&nbsp;
Fabiano Dalpiaz's Bio:
Fabiano Dalpiaz is a professor of software production at Utrecht University in the Netherlands. He is principal investigator in the department's Requirements Engineering lab. His research blends artificial intelligence with information visualization in order to increase the quality of the requirements engineering process and artifacts, with the ultimate aim of delivering higher-quality software. His research is often validated in-vivo through collaborations with the software industry. He acted as program co-chair for various conferences including RE 2023, REFSQ 2021, and RCIS 2020. He is an associate editor for the Requirements Engineering Journal and the Business &amp; Information Systems Engineering Journal, and he is the Steering Committee Chair of REFSQ. He often serves on the program committee of conferences such as RE, CAiSE, REFSQ, and ICSE.

Tuesday Oct. 29th, 10-11 am, via Zoom.
 
Title: Turning myself into an AI
Abstract: 
A discussion on what identity means when we have indistinguishable digital replicas and how we can achieve digital immortality through accurate AI representations of ourselves.
Dan Thompson's Bio:
Dan Thomson is founder of Sensay, a leading platform for the creation of AI-based digital clones and a pioneer in digital immortality. He founded Sensay in 2023 to create solutions across a range of sectors, from healthcare to education, building a knowledge capture and share platform for the betterment of humanity.He has a track record of launching and scaling new ideas, whether in the food and beverage industry or technology startups.Dan has a BA in Philosophy from King’s College London and an MBA from the University of Cambridge. He has authored two books to date: ‘Immortality in a Digital Age’ and ‘The Digital Afterlife’.

Sept 30th, 11-12 hybrid in Jupiter 473 and zoom.
&nbsp;
Title: "Prompting the Future: Integrating Generative LLMs and Requirements Engineering"
Abstract:
Generative LLMs, such as GPT, have revolutionized our interaction with artificial intelligence. Their ability to understand, generate, and manipulate language presents unprecedented opportunities and challenges across various disciplines, including RE. Generative LLMs have the potential to redefine the landscape of requirements elicitation, specification, and validation. The talk is structured into two primary segments. The first part delves into the application of generative LLMs in automating RE tasks. It explores how these models can assist in accurately capturing and specifying requirements, generating requirement documents, and automating the verification of requirements consistency and completeness. By examining examples and current research, this part will highlight the transformative potential of generative LLMs in enhancing efficiency, accuracy, and comprehensiveness in RE. The second part positions RE as a critical discipline for developing well-crafted prompts essential for interacting with generative LLMs. It underscores the importance of precise, unambiguous, and comprehensive requirements in formulating prompts that elicit accurate and relevant responses from the models. This part will also discuss the art and science of crafting effective prompts, drawing parallels between requirements specification techniques and prompt engineering strategies.
Bio:
Andreas Vogelsang is full professor for Software and Systems Engineering at the University of Cologne. He received a PhD from the Technical University of Munich. His research is at the intersection between requirements engineering, empirical software engineering, and machine learning. He has published over 100 papers in international journals and conferences such as TSE, JSS, IEEE Software, RE, and ICSE. In 2018, he was appointed as Junior-Fellow of the German Society for Informatics (GI). Further information can be obtained from https://cs.uni- koeln.de/sse.

Lab members

List of Topics

Contact the AI Engineering Lab

AI Engineering Lab

EN Black Chalmers GU Centr Plate

Choose subject

In order to play this video you must open your cookie settings and allow external media cookies.

In order to use basic functions like page navigation and access secure areas of the website, you must open your cookie settings and allow necessary cookies.

In order to use this service you must open your cookie settings and allow external services cookies.

To help us understand how visitors interact with the website you must open your cookie settings and allow statistics cookies.

Course syllabus for

Subject

Each academic year contains different courses

See programme page at universityadmissions.se

Educational area

Permissible substitutions of compulsory courses

To see your programme syllabus – select the academic year in which you started the programme.

[en: page.studyprogrammepage.syllabusesview.coursesdescription]

AI Engineering Lab

Recent and upcoming events

Upcoming Events:

Seminar and Workshop: Responsible and Safe Agentic AI Engineering

Seminar and Workshop: The Challenge of Designing High-Quality Green ML-Enabled Systems: A Software Architecture Perspective

Past Events:

WASP Software Engineering and Technology Cluster Workshop

Welcome to the next AI Engineering Lab's research seminar: AI Assistants for graded assignments

Seminar - Requirements Beyond Specifications: Use Cases for Modern AI by Fabiano Dalpiaz, 25th Feb 2025

Seminar - AI-based digital clones by Dan Thompson from Sensay

Seminar - Prompting the Future: Integrating Generative LLMs and Requirements Engineering by Andreas Vogelsang