ODSC London Data Science

Pre-registration is REQUIRED.
Add to your calendar - https://hubs.li/Q03lF-X-0

In this hands-on session, you'll learn how to generate high-quality synthetic data that preserves privacy using differential privacy techniques. We’ll walk through how to train differentially private generative models with MOSTLY AI’s open-source Synthetic Data SDK and explore how this method compares to traditional anonymization approaches in terms of both utility and risk. You’ll gain practical insights into configuring privacy parameters, understanding the impact of privacy budgets, and evaluating synthetic data output.
We’ll also cover how to assess the fidelity of synthetic datasets using predictive and discriminative machine learning models, and how to create hybrid datasets that blend real and synthetic data for improved utility. Through live demonstrations and real-world examples, you’ll develop a strong understanding of the privacy-utility trade-offs and how to confidently apply privacy-safe synthetic data in your own data science workflows.

Session Outline:
Lesson 1: Introduction to Differential Privacy Get familiar with the core concepts of differential privacy and how it differs from traditional anonymization techniques. By the end of this lesson, you’ll be able to explain what differential privacy is, what a privacy budget (epsilon) means, and why it provides stronger privacy guarantees than pseudonymization or masking.

Lesson 2: Setting Up and Using the Synthetic Data SDK Learn how to install and configure MOSTLY AI’s open-source Synthetic Data SDK to generate synthetic datasets with differential privacy enabled. You’ll run the SDK in LOCAL mode using a prepared dataset, explore the configuration options for privacy settings, and review the structure of the synthetic output.

Lesson 3: Evaluating Utility vs. Privacy Trade-offs Compare synthetic datasets generated with different privacy settings to understand how utility is impacted by stricter privacy budgets. By the end of this lesson, you’ll be able to evaluate the usefulness of differentially private synthetic data using predictive models and summary statistics.

Lesson 4: Creating Hybrid Datasets with Real and Synthetic Data Explore how to combine real and synthetic data to create hybrid datasets that retain utility while improving privacy. You’ll walk through a practical example and learn how to use synthetic data to augment or replace sensitive parts of your dataset.

Difficulty: Intermediate

Pre-reqs:
This tutorial is designed for data engineers, data scientists, ML engineers, and analysts with basic Python skills and familiarity with working in Jupyter Notebooks. Attendees should have a general understanding of machine learning workflows and working with tabular datasets (e.g., CSV files or pandas DataFrames). No prior experience with synthetic data is required. To participate fully in the hands-on exercises, attendees should have the following installed before the session: Python 3.11+, Git Installation and setup of the Synthetic Data SDK will be covered as part of the tutorial. But feel free to get started beforehand by visiting https://github.com/mostly-ai/mostlyai.

Speaker: Dr. Michael Platzer, Co-Founder and CTO of MOSTLY AI
Dr. Michael Platzer is co-founder and CTO of MOSTLY AI, a leader in privacy-safe synthetic data generation. He earned his degrees in mathematics and in business with distinction, led consumer analytic teams at global technology leaders, before starting his venture to pioneer the field of synthetic data. His company's mission is to democratize data access and data insights in a safe and responsible way for everyone.

ODSC Links:
• Get free access to more talks/trainings like this at Ai+ Training platform:
https://hubs.li/H0Zycsf0
• ODSC blog: https://opendatascience.com/
• Facebook: https://www.facebook.com/OPENDATASCI
• Twitter: https://twitter.com/_ODSC & @odsc
• LinkedIn: https://www.linkedin.com/company/open-data-science
• Slack Channel: https://hubs.li/Q038cQBy0
• Code of conduct: https://odsc.com/code-of-conduct/

Pre-registration is REQUIRED.
Add to your calendar - https://hubs.li/Q03kk6YJ0

AI agents are only as good as what they remember. Their ability to recall, reason, and act on relevant information is becoming a make-or-break factor. But building that memory—from raw data to structured, useful context—is still a major challenge.

In this webinar, we kick things off with a perspective on why this matters now. Drawing on experiences from data science and intelligent systems, we’ll highlight the growing importance of agent memory in real-world applications—and the role of knowledge graphs and contextual retrievers in making agents smarter, faster, and more trustworthy.

Then, we’ll dive into the how. Using Hypermode’s open-source tools, let’s walk through the process of building agent-ready context: crafting retrievers and feature stores, and designing tools that expose your graph in ways agents can actually use. We share the key design decisions behind these tools, including trade-offs, live demos, and lessons learned from real implementations.

What you’ll learn:
- Why memory is a core capability for powerful agents
- What it takes to expose graph data meaningfully to agents in live environments
- How to design and curate tools that align with your agents’ worldviews and goals

This session is great for those building their first AI agent and those looking to improve how agents reason and act.

Speaker#1: Kevin Van Gundy, CEO of Hypermode
He has spent his career building for developers, data scientists, and engineers. Previously he was the Chief Operating Officer and Chief Revenue Officer at Vercel and has held various leadership roles at Domino Data, Tray.io, and Neo4j. Kevin is involved in investing and advising startups through his involvement at Van Gundy Farms. His experience ranges from working in private equity to leading global go-to-market strategies for various companies. He has a Bachelor of Arts in English and Economics from the University of California, Davis.

Speaker2: William Lyon, AI Engineer at Hypermode
He works to improve the developer experience of building agentic flows. Previously he worked as a software developer at Neo4j and other startups. He is also the author of the book "Fullstack GraphQL Applications" and earned a masters degree in Computer Science from the University of Montana. You can find him online at lyonwj.com

ODSC Links:
• Get free access to more talks/trainings like this at Ai+ Training platform:
https://hubs.li/H0Zycsf0
• ODSC blog: https://opendatascience.com/
• Facebook: https://www.facebook.com/OPENDATASCI
• Twitter: https://twitter.com/_ODSC & @odsc
• LinkedIn: https://www.linkedin.com/company/open-data-science
• Slack Channel: https://hubs.li/Q038cQBy0
• Code of conduct: https://odsc.com/code-of-conduct/