Leadership changes at AGIBot!
Just now, QbitAI discovered that the name of former Chief Scientist Luo Jianlan is no longer present on the latest list of the partner team on the AGIBot official website.
In the previous official website list, Luo Jianlan's title was:
AGIBot Partner, Senior Vice President, and Chief Scientist.
When AGIBot first publicly disclosed its partner team in September 2025, Luo Jianlan appeared on the official website alongside Deng Taihua, Peng Zhihui, Yao Maoqing, and others, holding the title of Partner, Senior Vice President, and Chief Scientist.
In fact, even before this change to the official website list, there were many rumors on social media about Luo Jianlan's potential departure.
Now, with his name formally disappearing from the AGIBot website, this rumor seems to have gained another layer of corroboration.
Additionally, two other changes are worth noting.
Currently, Luo Jianlan's personal homepage only retains his affiliation as an Assistant Professor at Shanghai Chuangzhi College, with no mention of AGIBot in his personal bio.
Similarly, his X profile bio does not retain any positions related to AGIBot.
If these changes indeed correspond to an adjustment in his position, then it is highly likely that Luo Jianlan has already left AGIBot.
It is also noteworthy that the timing of this suspected personnel change is rather delicate.
Just recently, on July 24, 2026, AGIBot Innovation publicly confirmed that the company had initiated the process for a Hong Kong IPO.
However, as of the time of writing, neither AGIBot nor Luo Jianlan himself has publicly announced this personnel change. The specific details still await official confirmation.
Tenure of Approximately 1 Year and 4 Months
In April 2025, AGIBot officially announced Luo Jianlan's joining as Chief Scientist.
Simultaneously, the "AGIBot Embodied AI Research Center" he led in establishing was also formally founded, primarily responsible for cutting-edge embodied AI algorithm research and the engineering implementation of related technologies.
Subsequently, Luo Jianlan's position was further elevated to AGIBot Partner, Senior Vice President, and Chief Scientist, joining the company's core management and technical team.
If this departure news is true, from joining in April 2025 to leaving in August 2026, Luo Jianlan's formal tenure at AGIBot lasted approximately 1 year and 4 months.
Notably, just last week, Luo Jianlan was still promoting his work on VLA at AGIBot on X.
Overall, Luo Jianlan's research directions mainly revolve around robot learning, real-world reinforcement learning, and embodied intelligence. To date, his Google Scholar citation count has reached 8,449.
Looking back at Luo Jianlan's academic and research career is also quite impressive.
Luo Jianlan was born in 1993 and received his bachelor's degree from Wuhan University of Technology.
In 2015, he went to UC Berkeley for further studies and obtained his Ph.D. in 2020. During his time at Berkeley, he was advised by robotics luminaries Pieter Abbeel and Alice Agogino.
After his Ph.D., Luo Jianlan joined Google X as a researcher, collaborating with roboticist Stefan Schaal, and also participated in related research at DeepMind and Everyday Robots during his time there.
After leaving Google, he returned to the Berkeley Artificial Intelligence Research Laboratory (BAIR), working as a postdoctoral researcher in Sergey Levine's team, continuing research on real-world robot reinforcement learning.
There, he led the development of SERL and HIL-SERL.
Among these, SERL organized the algorithms, data collection, reward design, and robot control tools required for real-world reinforcement learning into a reusable system.
HIL-SERL further incorporated human intervention: robots explore autonomously most of the time, and only when they are about to fail or encounter difficulty do humans briefly take over. The system then uses these few but crucial corrective data points to continue learning, thereby reducing the reliance on a large amount of human demonstration in real-world reinforcement learning.
In April 2025, when Luo Jianlan officially joined AGIBot, he also brought this accumulated years of real-world reinforcement learning methodology into the company.
During his tenure of over a year at AGIBot, he was almost involved in building the entire research mainline for the company in real-world reinforcement learning, online post-training, and world models.
From promoting real-world reinforcement learning into industrial assembly scenarios, to using systems like SOP and LWD to connect robot deployment, data feedback, and model updates, and to incorporating future prediction and pre-execution evaluation into robot control through projects like τ0-WM, the representative works during Luo Jianlan's tenure, seemingly scattered, actually point towards the same goal:
Making the robot deployment process itself the starting point for the next round of learning.
Reference Links
[1]https://www.agibot.com.cn/company/leadership
[2]https://scholar.google.com/citations?hl=en&user=SJoRNbYAAAAJ&view_op=list_works&sortby=pubdate
This article is from the WeChat public account "QbitAI" (ID: QbitAI), author: Henry.
