Fewer Parameters, Fewer Problems - Precision AI with Small Language Models

Smaller, cheaper, faster, but somehow better?! Get hands-on with Small Language Models (SLMs) and discover how to deploy highly efficient, targeted AI as a real expert.

In this Build with AI workshop led by a skilled AI Solutions Engineer, you won’t just be looking at slides; you’ll learn by doing. We will be fine-tuning small language models and shipping them live via Ollama to tackle a real business problem. You'll master the nuances of training an on-device SLM and leave with practical knowledge on how to make smaller models punch above their weight. For some tasks we just don’t need generalists (LLMs), but we need focused specialists (SLMs).

Key Takeaways:

• Gemma 4: Working directly with Google's latest open models.

• Fine-Tuning: Customizing models for specific, real-world tasks.

• Live Build: Going from concept to a shipped solution in real-time.

• On-Device SLM: Deploying lightweight models for edge computing.

Who should attend: Developers, students, and architects looking to build, fine-tune, and scale the next generation of efficient, on-device AI applications.

What to bring:

• Your laptop (fully charged!).

• A curious open mind

• Optional: A basic familiarity with Python