talk-data.com

Pre-registration is REQUIRED. Add to your calendar - https://hubs.li/Q03HZj780

This hands-on workshop gives you experience taking your AI agent from prototype to production. First, you’ll build and deploy a secure MCP server on Google Cloud Run and implement authentication for authorized agents. Next, you’ll use the Agent Development Kit (ADK) to build and deploy an agent that uses the secure MCP server to accomplish its goals. We will also provide additional materials to show you how to deploy your agent to a GPU-accelerated Cloud Run instance for maximum power.

Speaker: Shir Meir Lador, AI Leader at Google

She is also the co-host of “The agent factory” podcast that dives into building production-ready AI agents. Previously, she worked as the AI group Manager of the Document Understanding Group at Intuit, where she led teams in developing AI services that helps consumers and small businesses prosper. Prior to intuit, Shir worked at 2 Israeli startups as a data scientist and researcher. A recognized leader in AI and data science, Shir has served as WiDS Tel Aviv Ambassador, KDD workshop organizer, co-founder and organizer of PyData Tel Aviv, and the co-host of “Unsupervised” a podcast exploring the latest in data science in Israel. She frequently speaks at international AI and data science conferences, sharing insights on applied machine learning and AI innovation. Shir holds an M.Sc. in Electrical Engineering and Computers from Ben-Gurion University, specializing in machine learning.

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.ai/code-of-conduct/

Pre-registration is REQUIRED. Add to your calendar - https://hubs.li/Q03HZj780

This hands-on workshop gives you experience taking your AI agent from prototype to production. First, you’ll build and deploy a secure MCP server on Google Cloud Run and implement authentication for authorized agents. Next, you’ll use the Agent Development Kit (ADK) to build and deploy an agent that uses the secure MCP server to accomplish its goals. We will also provide additional materials to show you how to deploy your agent to a GPU-accelerated Cloud Run instance for maximum power.

Speaker: Shir Meir Lador, AI Leader at Google

She is also the co-host of “The agent factory” podcast that dives into building production-ready AI agents. Previously, she worked as the AI group Manager of the Document Understanding Group at Intuit, where she led teams in developing AI services that helps consumers and small businesses prosper. Prior to intuit, Shir worked at 2 Israeli startups as a data scientist and researcher. A recognized leader in AI and data science, Shir has served as WiDS Tel Aviv Ambassador, KDD workshop organizer, co-founder and organizer of PyData Tel Aviv, and the co-host of “Unsupervised” a podcast exploring the latest in data science in Israel. She frequently speaks at international AI and data science conferences, sharing insights on applied machine learning and AI innovation. Shir holds an M.Sc. in Electrical Engineering and Computers from Ben-Gurion University, specializing in machine learning.

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.ai/code-of-conduct/

Pre-registration is REQUIRED. Add to your calendar - https://hubs.li/Q03HZj780

This hands-on workshop gives you experience taking your AI agent from prototype to production. First, you’ll build and deploy a secure MCP server on Google Cloud Run and implement authentication for authorized agents. Next, you’ll use the Agent Development Kit (ADK) to build and deploy an agent that uses the secure MCP server to accomplish its goals. We will also provide additional materials to show you how to deploy your agent to a GPU-accelerated Cloud Run instance for maximum power.

Speaker: Shir Meir Lador, AI Leader at Google

She is also the co-host of “The agent factory” podcast that dives into building production-ready AI agents. Previously, she worked as the AI group Manager of the Document Understanding Group at Intuit, where she led teams in developing AI services that helps consumers and small businesses prosper. Prior to intuit, Shir worked at 2 Israeli startups as a data scientist and researcher. A recognized leader in AI and data science, Shir has served as WiDS Tel Aviv Ambassador, KDD workshop organizer, co-founder and organizer of PyData Tel Aviv, and the co-host of “Unsupervised” a podcast exploring the latest in data science in Israel. She frequently speaks at international AI and data science conferences, sharing insights on applied machine learning and AI innovation. Shir holds an M.Sc. in Electrical Engineering and Computers from Ben-Gurion University, specializing in machine learning.

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.ai/code-of-conduct/

Pre-registration is REQUIRED. Add to your calendar - https://hubs.li/Q03HZj780

This hands-on workshop gives you experience taking your AI agent from prototype to production. First, you’ll build and deploy a secure MCP server on Google Cloud Run and implement authentication for authorized agents. Next, you’ll use the Agent Development Kit (ADK) to build and deploy an agent that uses the secure MCP server to accomplish its goals. We will also provide additional materials to show you how to deploy your agent to a GPU-accelerated Cloud Run instance for maximum power.

Speaker: Shir Meir Lador, AI Leader at Google

She is also the co-host of “The agent factory” podcast that dives into building production-ready AI agents. Previously, she worked as the AI group Manager of the Document Understanding Group at Intuit, where she led teams in developing AI services that helps consumers and small businesses prosper. Prior to intuit, Shir worked at 2 Israeli startups as a data scientist and researcher. A recognized leader in AI and data science, Shir has served as WiDS Tel Aviv Ambassador, KDD workshop organizer, co-founder and organizer of PyData Tel Aviv, and the co-host of “Unsupervised” a podcast exploring the latest in data science in Israel. She frequently speaks at international AI and data science conferences, sharing insights on applied machine learning and AI innovation. Shir holds an M.Sc. in Electrical Engineering and Computers from Ben-Gurion University, specializing in machine learning.

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.ai/code-of-conduct/

🔔 PyData Yerevan invites you to its May meetup!

Mohamed Rashad, Co-Founder & Engineering Lead at Hyperion, will present a talk on “Taking AI Agents from Prototype to Production.”

Building and deploying AI agents in real-world environments is no easy task — especially within the constraints of corporate infrastructure. In this talk, Mohamed will share best practices and architectural patterns for developing on-prem AI systems, with a focus on LLMs. He’ll cover how to navigate challenges related to security, compliance, networking, and software layers — drawing on hands-on experience across five countries, including Armenia.

📅 Date: May 22, 2025 🕖 Time: 19:00 📍 Location: 414W, PAB, American University of Armenia 🔗 Register here: https://forms.gle/Pn9wpwVMg8zWGknL7

📣📣 Attention Airflow Users and Big Data Professionals 📣📣

Come join us at Ramp's offices for a special meetup on the power of AI + Airflow!

***PLEASE NOTE, VERY IMPORTANT*** To attend this meetup, please register here. Registration for this meetup will not be open on this page.

AGENDA

• 6:00 PM-6:30: Eat, Drink and Network!
• 6:30 PM-8:00 PM: Presentations!
• 8:00 PM- 8:30 PM: Eat more, drink more, and network!

SPEAKERS Julian LaNeve, CTO at Astronomer From Prototype to Production: Solving Data Engineering Challenges with Airflow in Ask Astro

• Prototyping an LLM application is easy, but taking it to production is a lot more challenging. You'll face many typical data engineering challenges, which is exactly what Airflow was designed to solve! Join Astronomer as they talk about Ask Astro, an open-source question-answering LLM application across a variety of document sources. You'll learn how to create a retrieval augmented generation (RAG) application, including data ingestion and storage, prompt orchestration, and feedback loops.

Josiah Parappally, Software Engineer at Ramp Maximizing Customer Experience with LLM: Innovations in Support Ticket Onboarding

• Josiah Parappally will present challenges faced by new customer experience agents during onboarding, driven by the complexity of support tickets. He'll discuss strategies like using summaries, macros, and Airflow as an LLM tool, along with OpenSearch for categorization. The talk aims to enhance agent onboarding by streamlining support ticket resolution through innovative tools and techniques.

Sam Braun, Member of the Technical Staff at Cohere Co:lineage: Tracking Metadata and Lineage across LLM Training Datasets

• LLMs use datasets from a wide variety of sources in their training runs. How can we keep track of the metadata, lineage, and lifecycle information associated with these datasets? Sam will walk through co:lineage, a library built at Cohere to answer these questions. He'll discuss why we use both an event-driven architecture and Airflow for computing and collecting this metadata, and how we use Airflow jobs to monitor the integrity of our metadata store.