Panel Software Engineering in the age of AI
summary
A live panel surveys agent tooling and real projects: hands‑on use of LangChain4J and ADK for Java, greenfield work produces usable code quickly while legacy air‑gapped defence systems require local open models and months of tuning, maintainers spend half their time reviewing AI‑generated PRs, companies report up to 75% AI‑generated code internally, human review becomes the scaling bottleneck, guardrails and human‑in‑the‑loop approvals prevent autonomous changes, and statically typed Java-style APIs help models verify and compile outputs.
AI-written summary · about
chapters (10)
- 0:00Welcome and introductions
- 1:07Panelist backgrounds
- 4:20Code quality with AI agents
- 12:09Guardrails for AI coding
- 19:39Preview models in big tech
- 27:06Programming languages in AI era
- 36:10Trust in AI-generated code
- 39:01Future of open source
- 41:48Choosing AI models daily
- 43:34Sustainability and AI responsibility
related talks
56mPANEL: Unleashing the Future: The Impact of Gen AI on Development Productivity
46mPanel Discussion: AI's Impact on Developers
41mPanel discussion: AI in code development – hype vs reality
47mAI and the Future of Software Development
01h 05mPanel: Agentic SE & Open Source in the FMware Era – Toward Trustworthy, Context-Aware FMware
30mBuilding Tools with Tools: Modern AI-Driven Development
26mAI Agents as Your Co-Developer
50mPanel: Beyond The Vibe: Code Quality First