Well, the answer to that is up to you and how much time you want to put in. But to give you a sense, here are some demos from the projects some students in cohort 1 did:

• Tripify
• Interview Prep Agent
• E-commerce Agent

Per week:

• 2 hrs - live sessions
• 2-4 hrs - The reading / watching material
• 0-n hours of project

Summary. You need 4 hours a week to get the absolute min. out of this course. The upper bound is however much time you want to spend on the project or digging deeper into the material.

You have access to course material indefinitely.

We will be collecting questions and resources that will help you answer this question here. If you feel you know more than 30% of these concepts, and can learn another 30% by the time the course starts, then you're at the right place.

Full refund (minus fees) is available up to the course start date.

See this link.

Yes!

• If you are an AISC member, check Slack.
• If you are part of the focus group, check your email.
• If you want to be part of the above, send an email to maven@ai.science

If you are in a situation where enrolling in this course without a large discount is unaffordable for you, send me an email and we will address that on a case by case basis.

The primary purpose of the live sessions is for you to ask your questions. So, it is strongly encouraged that you arrange your schedule for the few weeks of the cohort to join those. We will try our best to provide the recordings of guest lectures and summaries of the questions and answers available.

That depends on how involved you want your project to be. We will provide guidance and templates for 3 levels of hands on experience:

• Strategy and architecture (eg. Miro)
• Quick prototyping (eg. Vertex AI Agent Builder)
• Coding (eg. Autogen)

There will be teaching assistants that will answer your technical questions, to the best of their ability, during the course and via Slack. After the course, you can stay in the same Slack workspace and post your technical questions and the course alumni and the broader community can answer.

Students will be encouraged to form groups of 2-3 and work on the projects together. Students will be prompted to provide intros and suggestions about what they want to work on via Slack, and that way they can meet each other and coordinate. Teaching assistants can also help refine the ideas.

Agents: LLM systems that have working memory, ability to use tools, and can plan their work.

Multi-agent Systems: Architectures that involve multi agents like above.

Some phrases you might have heard that are related:

• Agentic workflows / pipelines
• Cooperative / collaborative / collective AI
• Multi-player, human-machine interfaces

Yes. See this link for more details.