airhacks

In-person only. Register here 👉 https://www.meetup.com/java-vienna/events/314639734/

Modern Java provides all the functionality required for local LLM inference and agent orchestration, eliminating the need for external dependencies.
LightMetal directly binds Llama.cpp from Java using the Foreign Function and Memory API (FFM), eliminating the need for JNI or native wrappers. It runs open models such as Mistral, Gemma, and Llama on your machine.
Zsmith is built on LightMetal and uses the Boundary-Control-Entity (BCE) model to structure agents. It loads tools and skills as ServiceLoader plugins and runs tool calls and streaming on Virtual Threads. Local inference and remote models work with the same agent code over the JDK's HttpClient.

Topics:
- One-line installation, no build tools, no package managers
- FFM in practice: calling native libraries without JNI
- Running Mistral, Gemma and Llama locally in pure Java
- ServiceLoader as a plugin mechanism for tools and skills
- BCE applied to AI agents
- Virtual Threads for streaming, tool calls and parallel inference
- Switching between local and remote models without changing the agent
- Plain JDK: HttpClient, records, sealed types and pattern matching
- Packaging agents as single-file Java applications

Everything is demonstrated live with code. Your questions are highly welcome — the more, the better!

See you at: https://airhacks.live and https://airhacks.university!

# Join Adam Bien for a hands-on workshop on building modern, zero-dependency web applications with Web Components, scalable / feature-oriented architecture, and LLM assistance. From screenshot to deployable SPA — no frameworks required.

80% hands-on coding — the workshop adapts to your questions!

• From screenshot / mockup / wireframe to maintainable UIs
• Building SPAs with web standards and LLMs
• Effective prompting strategies for web components
• Iterative refinement of LLM-generated components
• Why semantic markup and BCE reduce LLM hallucinations
• Creating Domain Specific Prompts (DSPs) for frontend
• Taming large applications with feature-oriented structure
• Structuring feature modules with Boundary-Control-Entity
• Building high performance apps with Web Components and web standards
• Zero-dependency web apps with lit-html and Redux Toolkit
• Using Web Baseline to guide LLM technology decisions
• Capturing best practices with instructions, SKILL.md and AGENTS.md
• Implementing policies, guardrails and hooks
• Using specialised sub-agents for quality
• Import maps and ES modules without a build step
• Unidirectional data flow with Redux Toolkit
• Client-side routing without frameworks
• Custom events for component communication
• Automatic state persistence with localStorage
• Semantic HTML over div/span soup
• CSS-only interactivity: accordions, modals, tabs without JS
• Accessibility with WCAG and focus management
• Modern CSS: nesting, logical properties, clamp() and oklch()
• Dark mode with prefers-color-scheme
• Responsive design with CSS Grid and container queries
• Testing web components with Playwright and MCPs
• Cross-browser E2E testing with Playwright
• Effective GitHub / LLM integrations
• Working with local LLMs
• Working with Claude Code, Kiro, GitHub Copilot and (...)
• Effective backend integration and full stack development
• From prompt to deployable web component SPA
• ...(TBD)

Checkout for details: https://workshops.adam-bien.com/llm-assisted-webcomponents.htm