Software & AI Nights

No lectures, no fixed curriculum. Just people sharing what they’ve tried, what broke, and what surprised them. The format is deliberately light because the field moves faster than any one person can track alone — the point is collective sense-making.

How an evening runs

• Check-in — new people introduce themselves; everyone says what they’ve been trying, building or reading about since last time.
• Lightning demos and stories — 5 to 10 minutes each. Someone shows a tool, shares a workflow, or tells the story of something that worked, or didn’t.
• Open discussion — reactions, questions, and working out together what any of it means.
• Next session planning — we pick the topic or format for next time as a group.

Who it’s for

Developers, students and researchers who actually use agentic coding tools, from Softwarepark Hagenberg and beyond. You don’t need to be an expert. You need to have tried something and be willing to talk about it.

The focus is applied and practical: maximising real value for the people in the room, whether that’s a personal project or a business problem. Academic and research perspectives are welcome as long as they stay grounded in something we can try together.

The bigger picture

The immediate focus is day-to-day coding assistants, but this feeds a broader horizon. The biggest limitation of the exploding LLM market is a lack of correctness. By experimenting with agentic orchestration, tool use and standard protocols now, we’re exploring the integration layers needed for a much bigger breakthrough in generating correct code. We’re not just keeping up with the state of the art — we’re feeling out what comes after it.

How it’s organised

Responsibility rotates. There is no single permanent organiser; someone volunteers or is picked to own each session. Members can bring friends and colleagues, though warm introductions are preferred. Sessions are in person only. Notes and resources are shared afterwards so nothing gets lost.