Tutorial Description

At SANER 2024, we focus on the significant potential of AI-agent-based methodologies in software engineering through a focused exploration of LLM Agents and novel models from the GPT lab at Tampere University. This tutorial, led by GPT-lab experts from Tampere University and in the application of large language models (LLMs) in the software engineering domain offers an exciting learning journey into the automation of complex software development processes using AI agents.

Learning Objectives:

Participants will

-Understand the theoretical underpinnings of employing LLMs and GPTs in automating software development. Gain hands-on experience with the latest agent LLM agent models, and their application in generating, reviewing, verifying, and testing code.

-Learn to apply AI-agent-based strategies to real-world software projects, enhancing efficiency, accuracy, and security.

Interactive Sessions and Case Studies

Engage in live coding sessions where you will use LLM Agent models to generate running code from natural language prompts, address even complex software project challenges, and evaluate the outcome against industry benchmarks. Interactive workshops will facilitate deep dives into specific case studies, showcasing the successful application of LLM. Agents in software development projects, highlighting its efficiency, cost-effectiveness, and ability to mitigate security vulnerabilities.

Outcomes and Takeaways

Attendees will leave the tutorial equipped with:

-New skills in deploying AI agents for software development tasks.

-Exclusive access to LLM Agent models, including insights into its operational framework and application protocols.

An understanding of the latest research and practical applications of LLMs in software engineering, ready to be applied in their projects for improved efficiency and innovation.