Agents Week

powered by the Yandex School of Data Analysis

  • Yandex School of Data Analysis experts explain the architecture of modern AI agents, covering the full process: building agents from scratch, configuring their behavior, connecting external systems, and scaling solutions for real workloads.

  • The video sessions are over, but all the video recordings are available on YouTube.

From LLM prompts
to agent systems

  • Who is the intensive for?

    This intensive is designed for anyone planning to implement agent systems in real products or infrastructure, or build personal assistants:

     

    • ML engineers, backend developers, and researchers who want to move from isolated LLM prompts and prototypes to full agent systems.
    • Technical university students with strong coding skills eager to understand AI agents.
  • What will it cover?

    • How modern AI agents work: from LLMs and prompts to the thought—action—observation loop, tools, memory, and agent architectures.
    •  

    • Designing and configuring agent behavior: reasoning, orchestration of multi-agent systems, and guardrails, with real engineering constraints.
    •  

    • Evaluating AI agents: сore approaches to assessing solutions and checking the reliability and stability of agent behavior.
    •  

    • Production engineering principles for AI agents: setting up monitoring, scaling, and maintaining systems under real workloads.

What’s in the program

06.04
Theme 1.1. Intro to AI Agents LLM
Alyona Zaitseva

Leads the AI services team at Yandex Lavka, developing customer-facing and internal AI agents and the platform that supports them

Watch on YouTube
07.04
Theme 1.2 
Tools. MCP
Alyona Zaitseva

Leads the AI services team at Yandex Lavka, developing customer-facing and internal AI agents and the platform that supports them

Watch on YouTube
08.04
Theme 2 
Memory and Guardrails in LLM-Powered Agents
Kirill Mishchenko

Leads the ML development team in Yandex Browser

Watch on YouTube
09.04
Workshop 2
. Memory and Guardrails in LLM-Powered Agents
Kirill Mishchenko

Leads the ML development team in Yandex Browser

Watch on YouTube
10.04
Theme 3.1 AI Agent Workflow Multi-Agent Systems Multimodality
Sofya Proskurina

Works on the platform for internal AI agents and develops agents that help optimize developer workflows in Yandex Lavka

Watch on YouTube
11.04
Theme 3.2
. AI Agent Workflow Multi-Agent Systems Multimodality
Sofya Proskurina

Works on the platform for internal AI agents and develops agents that help optimize developer workflows in Yandex Lavka

Watch on YouTube
12.04
Theme 4. Agent Evaluation: From Metrics to Managed Quality
Sergey Kuptsov

Develops agent solutions for Alice and Smart devices

Watch on YouTube
13.04
Workshop 4. Agent Evaluation: From Metrics to Managed Quality
Sergey Kuptsov

Develops agent solutions for Alice and Smart devices

Watch on YouTube
14.04
Theme 5.1. Production Engineering for LLM Agents
Daniil Artamonov

Is responsible for the platform of internal AI agents and the agents that accelerate development, testing, and product launches in Yandex Lavka

Watch on YouTube
15.04
Theme 5.2. Production Engineering for LLM Agents
Kirill Vlasov

Leads the development of AI Studio Yandex Cloud

Watch on YouTube

FAQ

Still have questions?

Contact us —

we will be happy to help.