Building & Governing AI: Semantic Search with DynamoDB + EU AI Act Risk Tiers

Agenda:
18:30 - 18:40 hrs - Welcome , Community updates and Intro

18:40 - 19:10 hrs :
Topic 1: One Less Database to Operate: Rethinking Semantic Search with DynamoDB - [Host: Chidimma Dominic Nweke, AWS Certified Software Engineer]
Adding semantic search often means adding another datastore, replication pipeline, and operational burden. But what if your existing database could handle it too?
In this session, we’ll build a semantic search application using Amazon Bedrock embeddings and DynamoDB vector search, exploring how vector search fits alongside traditional DynamoDB access patterns.
We’ll also look beyond the demo at the architectural trade-offs, from cost and observability to coupling and consistency, and discuss when DynamoDB is enough and when a dedicated vector database is still the better choice.
You’ll leave with a practical way to think about adding semantic search without automatically adding another distributed system to operate.

19:10 - 19:15 hrs Q&A

19:15 - 19:25 hrs: 10 min fun event(Quiz)

19:25 - 19:55 hrs:
Topic 2: EU AI Act Demystified: Risk Tiers, August 2 Rules, and Future Requirements - [Host: Hycentina Ifeoma Ekwerike, Certified AI Governance Professional (AIGP), Founder, Cinthify]
The EU AI Act sorts AI systems into four risk tiers: unacceptable, high, limited and minimal. Which tier your system falls into decides what you have to do, so it’s the first thing to work out. In this 30-minute session I’ll go through each tier with the kinds of systems that land in it. That covers what’s banned outright, what counts as high-risk (recruitment tools and credit scoring are the usual examples), what you have to disclose to users, and what sits at minimal risk. I’ll also cover where people get it wrong, including how the same tool can land in a different tier depending on what you use it for. Fines under the Act go up to €35 million or 7% of global annual turnover for the most serious breaches, so this is worth getting right early. This is Part 1 of a series on AI governance. [LATER TOPICS, e.g. what high-risk obligations look like in practice] Who it’s for: founders, product and engineering leads, and legal, compliance and risk professionals. You don’t need any prior knowledge of the Act.

19:55 - 20:00 hrs: Q&A , Next episode updates