[Alibaba DAMO Academy × Datawhale] 5-Day Embodied AI Practice Camp! Join leading trainees X to unlock the “body” and future of AI

This article is a draft for the public account recruitment copy. Before the official release, please replace the start date, participant introduction, registration link, QR code, and number of recruits.

Hello, all my learning partners from Datawhale.

Are you curious about the cutting-edge field of embodied AI, yet overwhelmed by complex concepts and unclear pathways, leaving you at a loss? Do you want to understand how AI can gain a “body” for perception and action from its powerful “brain,” so it can comprehend tasks, plan actions, and engage in interactions in the real physical world?

Now, a quick beginner-friendly guide is ready.

We plan to launch the Alibaba DAMO Academy × Datawhale “Embodied AI for Beginners: 5-Day Practical Course”. This learning session will focus on the embodied AI topics jointly organized by Alibaba DAMO Academy and Datawhale. Led by participant X, everyone will complete a systematic cognitive upgrade in just 5 days.

This is not a complex paper reproduction, nor does it immediately dump everyone into complex environment setups. It’s more like an efficient tour: first, it draws a panoramic map of embodied AI for you; then, it helps you understand mainstream approaches such as VLA, world model, and RynnBrain; finally, through the DaMo Academy LeCloud platform, you can experience a closed loop from idea to simple implementation.

Your Navigator: X

Leader: X

This 5-day systematic learning session will be led by X throughout. As a “guide” and “companion,” X will assist you in completing the entire closed loop process, from building understanding to developing personal ideas.

During daily learning sessions, X will provide guidance notes based on the thematic materials, outline the key points of the day, collect questions from everyone, and organize group Q&A sessions. You don’t need to master robotics, reinforcement learning, or real robot deployment from the start. More importantly, you need to keep up with the pace, break down complex concepts, and document your daily insights.

In 5 days, what will you systematically master?

We will focus on the topics in the Dharma Institute within Every Embodied. This material has a clear structure, making it suitable for team-based learning: one specific theme per day, culminating in a personal application concept that can be submitted, shared, and further improved.

Your 5-day learning goal is very clear:

• Establish a comprehensive cognitive map: Clarify what problems embodied AI is solving, and its relationship with large models, robot control, and agents.
• Inspect mainstream technical approaches: Understand the differences and connections between routes such as VLA, VLM, world model, and RynnBrain.
• Experience the closed loop in platform implementation: Observe how data, models, skills, and deployment are interconnected on the DaMo Academy LeYun platform to gain a practical sense of engineering.
• Develop personal application concepts: Organize fragmented information into a structured understanding, and ultimately produce a personalized “embodied AI application concept document”.

Core learning path and resources:

• Open-source course entry: Every Embodied
• Special topic learning directory: 16-Special Team Learning/01-Damo Institute Team Learning
• Practice platform: Damo Institute Music Cloud Platform

5-Day System Learning Schedule

Instructor X will guide everyone through the process day by day, ensuring that daily goals are clear, learning content can be completed, and final outcomes can be implemented.

Day 0: Opening Ceremony and Cognitive Warm-up

X will conduct an opening sharing session to introduce learning objectives, learning rhythm, data structure, and final output requirements. Everyone needs to prepare their GitHub accounts, platform access links, and study materials, and write down the most interesting embodied AI application scenarios that interest them.

Day 1: Theory and Trends

Following Task 01《Theory and Trends》, understand the definition, development drivers, typical product forms, and industry challenges of embodied AI. The focus today is on creating the first cognitive map: what embodied AI is, and why it becomes important at this point in time.

Day 2: Technical Analysis

Enter Task 02 “Technical Perspective” to systematically review core concepts such as VLA, VLM, world model, RynnBrain, and RynnRCP. X will further refine key issues like the “visual-language-action model”, “world model”, and “the relationship between platform capabilities and model capabilities”.

Day 3: Platform Demonstration and Practical Experience

Enter Task 03: “Demonstration Practice: From Idea to Implementation”. Under the guidance of the instructor, log in to the DaMo Academy LeCloud Platform and observe the model marketplace, data management, model services, skill debugging, and deployment processes. The goal for today is not to achieve complex implementations, but to understand manually the basic workflow of an embodied AI task from concept to system.

Day 4: Summary, Sharing, and Concept Output

Complete Task 04: “Summary and Sharing”. Organize the learning insights from the past few days to create your own “Personal Application Blueprint”. This blueprint should cover scenarios such as desktop organization, campus navigation, household shopping, laboratory supplies sorting, and auxiliary operations. The focus is on clearly outlining tasks, inputs, actions, routes, and challenges.

Why is the 5-day learning session worth participating in this time?

1. Jointly produced by top institutions

The learning materials focus on the embodied AI topic jointly organized by Alibaba DAMO Academy and Datawhale. They include cutting-edge concepts as well as platform-based practice resources, making it suitable for beginners to quickly establish their understanding in the field.

2. Friendly for beginners with clear paths

No prior knowledge of robotics, reinforcement learning, or real robotic arm manipulation is required. The 5-day program focuses on 4 core tasks to quickly build a comprehensive understanding of the field, rather than getting stuck on environment setup from the start.

3. The instructor will guide the students throughout the learning process

The instructor X will provide daily guidance, key point extraction, question collection, and centralized Q&A. The greatest value of group learning is that the knowledge points that one person may get stuck on can be discussed and advanced together by everyone.

4. Theory combined with real platform experience

This learning session not only reviews cutting-edge approaches such as VLA and world model, but also allows us to observe the data, model, skill debugging, and deployment processes through the DaMo Academy LeCloud platform, avoiding staying at the stage of merely “reading concepts”.

5. Harvest structured outputs that can be archived

The final “Personal Application Concept Document” is the result of this learning experience. It is not required to be written as a paper, but it needs to clearly describe a specific scenario: who will use it, where it will be used, what actions need to be performed, what technical approaches might be used, and what the most challenging aspects will be in the future.

What support will you receive?

• Lead Learner X: Responsible for the learning pace, key explanations, problem collection, centralized Q&A, and summary sharing.
• Datawhale Community: Provides team-based learning organization, material access maintenance, and learning process support.
• Every Embodied Open Source Tutorial: Offers continuously accessible learning materials, allowing deeper exploration after the event.
• Peer Learning Environment: With clear daily goals, the learning process is easier to maintain, and problems can be explained more clearly.

Suitable for such learners

• Interested in cutting-edge technologies such as AI, robots, embodied AI, and agents, and eager to build a systematic understanding quickly;
• Tired of fragmented information, wanting to complete a focused learning sprint in 5 days;
• Want to first understand the relationship between VLA, world model, and platform practices before deciding whether to delve deeper into code, simulation, or real robot development;
• Prefer clear guidance, and willing to share my understanding during group discussions.

Get on the train now, and unlock a new world of embodied AI in 5 days

We plan to recruit [Number of Recipients to Be Determined] learners to jointly launch this intensive, clear, and future-oriented introductory course on embodied AI.

Scan the QR code below to join the exclusive learning group for “Alibaba DAMO Academy × Datawhale Embodied AI 5-Day Practical Camp”.

[QR Code]

Scan the code to join the group, secure your seat, and join the trainees X to explore the new world of embodied AI in 5 days!

As a team-based learning activity, we believe that what is more important than acquiring fragmented knowledge is the ability to quickly and systematically establish a cognitive framework for a particular field. We look forward to working with you to open the doors to the embodied AI world during these 5 enriching days, understand its intricacies, and find our own direction for exploration.