Viktoria Semaan

Not every problem needs the biggest model — and not every model works out of the box. With frontier models, small open models, and fine-tuning all on the table, the hardest question isn't "can we do AI?" It's "which model is actually right for this?"In this session, we walk through how to match models to use cases across real enterprise workloads. You'll see when a frontier model earns its cost, when a small open model wins on latency, and when fine-tuning turns a generic model into a specialist that knows your business. We'll put the tradeoffs in action — quality, cost, latency, control — using concrete examples across customer support, document processing, and analytics.You'll leave with a repeatable framework for choosing, customizing, and deploying the right model on Databricks every time — and the patterns to keep those decisions sharp as new models land.

Viktoria Semaan walks you through two ways to build data apps on Databricks, using a real-world use case: extracting insights from public company 10-K filings, storing structured financial data in Databricks, and running analytics on it immediately. The first path uses Claude Code with the Databricks AI Dev Kit, the production-grade approach for enterprise developers who want full control, governance, and deep Databricks integration. The second path uses Lovable, the natural fit for vibe coding, rapid prototyping, and getting non-developers building fast. Both connect directly to Databricks. You will leave with a clear view of the trade-offs between the two approaches, the resources to get started the same day, and a new benchmark for how fast intelligent data apps can be built and deployed.

In this session, discover how AI services, including Amazon Q Business, can help to scale and improve community engagement, streamline events planning, and handle everyday tasks as an event organizer. Dive into technical insights with a demo as we share practical ideas and offer guidance on getting started.

As generative AI (gen AI) transitions to production systems in 2024, early adopters are reaping rewards like accelerated productivity and customer experiences. Organizations must build customized gen AI models and applications aligning with their use cases and culture.

Deriving insights from data can be challenging due to inefficient data preparation, cleansing, exploration, and visualization. In this session, learn how the capabilities of Amazon SageMaker, coupled with Amazon CodeWhisperer, can revolutionize your data preparation and data management workflows.

Operational data is notoriously challenging to unlock, a task complicated by disparate equipment, sensors, device protocols, and sub systems and the need to securely connect global networks. Yet, doing so activates the remote health monitoring of industrial assets.