Frontend Developer Meetup Amsterdam

Applied AI Development for TypeScript Engineers Workshop

Duration: 2.5 hours
Prerequisites: TypeScript, Node.js fundamentals

Doors Open: 18:00
Workshop Start: 18:30

Topics:

• An Introduction to Large Language Models
• Building with the Vercel AI SDK
• AI Tooling and DX
• Vector Embeddings & Semantic Search
• Privacy-First AI Development

Workshop Overview
Transform your TypeScript/Node.js applications with practical AI integration techniques. This hands-on workshop focuses on real-world implementations you can use immediately. We'll also tackle privacy issues.

18:30-18:45 - An Introduction to Large Language Models
Dive into the architecture of transformer-based models without getting lost in mathematical complexity. We'll break down the key concepts that matter for practical implementation, focusing on attention mechanisms and how they enable the contextual understanding that powers modern AI.

18:45-19:15 - Building with the Vercel AI SDK
Master the Vercel AI SDK to create type-safe AI features in your applications, including structured JSON outputs and streaming responses. You'll implement real-time text generation, object generation, and data transformation while maintaining type safety throughout your application. You'll also learn how to change LLM providers in a few simple steps.

19:15-19:30 - AI Tooling and DX
Unlock the true potential of AI in development with a deep dive into the latest cutting-edge AI tools transforming the industry. Experience a live demonstration of Cursor IDE, an AI-powered coding assistant that reduces hallucinations by allowing you to ingest both local and remote documentation as reference material for coding assistants.

19:30-20:30 - Vector Embeddings & Semantic Search
Learn to implement semantic search and similarity matching using vector embeddings in TypeScript applications.
This alternative to keyword-based querying allows for powerful semantic search throughout your web application.

20:30-21:00 - Privacy-First AI Development
Explore techniques for running AI models locally using Ollama. We'll implement a private AI assistant that processes sensitive data entirely on-premise while maintaining high performance.

21:00-21:15- Q&A