Room 1 - Keynote / Breakout

ip.labs is the world’s leading white label e-commerce software imaging company and processes millions of images every day. The workflows of our users consist of designing, saving, loading, ordering and delivering to the printing facilities the photo products like prints, photobooks, calendars, gift products among others. In this talk we’ll explore our motivation and architecture behind the reimplementation of our image storage solution based on AWS Serverless services like API Gateway, Lambda, DynamoDB, SQS, SNS, EventBridge, Kinesis and others. We’ll especially dive deeper into the parts using event-driven communication and explore patterns like storage-first, fan out, change log and cross-account logging among others and explain our architectural decisions.

This session explores the Model Context Protocol (MCP) from basic concepts to practical applications. We’ll start with generative AI fundamentals and quickly move to how MCP enhances AI systems through structured context handling.

We’ll examine MCP’s technical architecture and how it improves conversations with AI by enabling RAG systems and agent workflows. The presentation breaks down MCP’s key components and integration points for developers looking to implement this technology.

The talk concludes with live demonstrations of MCP in action: knowledge retrieval using Amazon Bedrock, Kubernetes troubleshooting automation, and Terraform code generation. Attendees will learn how to leverage MCP to build more capable AI applications across various technical domains.

In complex data platforms powered by AWS Glue, AppFlow, Airflow, and Step Functions, operational issues are inevitable. This talk shares how we built a self-healing system that automatically detects issues, tags them with context, uses Bedrock-based LLMs to suggest resolutions, and keeps stakeholders informed via GitLab and Slack. The result: a 40% drop in incident volume and significantly faster resolution times.

Key takeaways: • Using automation + AI for smarter incident workflows • Practical use of Bedrock in a data engineering context • Bridging operations and business with structured communication

This session is about pushing boundaries with AWS hybrid architectures. We’ll explore how to combine Raspberry Pi nodes, EKS, DMS, and generative AI for smarter data transfer and flexible hybrid setups. It’s a practical and innovative way to handle complex workloads on a budget.

Search engines guide users from their initial goals to the information that fulfills those goals. The retrieved items are deemed relevant when they align with the user’s intent. Ensuring relevant results involves having the data in the catalog, analyzing it accurately, representing it through text and vectors, ranking it appropriately, and performing other actions. In this session, discover how auction houses utilize vector search, powered by Amazon OpenSearch Service, to find similar items within their extensive database, enhancing the customer experience and preventing issues with counterfeit artworks, and see how it assists users in obtaining the most relevant results. This session also uncovers some peculiarities in getting embeddings with AWS Bedrock and AWS SageMaker serverless inference for the abovementioned goals.

What happens when AI makes code writing the easiest part of development? The rise of Generative AI has transformed software development, making rapid prototyping more accessible than ever. This keynote examines the evolving role of developers as AI reshapes traditional coding practices. The presentation cuts through both hype and fear to explore what truly matters in modern software development. Learn why moving from intuition-based ‘vibe coding’ to structured engineering approaches becomes more crucial with AI assistance. The session demonstrates how specification-driven development and engineering principles create the foundation for successful AI collaboration.