- Co-founder and Chief AI Officer, All Hands AI (OpenHands)12
- Computer science PhD candidate advised by Heng Ji, University of Illinois Urbana-Champaign1
Xingyao Wang is a co-founder of All Hands AI, which builds the open-source software agent OpenHands, and is listed on the company's team page as Chief AI Officer. Wang is also a computer science PhD candidate at the University of Illinois Urbana-Champaign advised by Heng Ji. Wang is first author of the OpenHands paper, the CodeAct paper on executable code actions for LLM agents, and the OpenHands Software Agent SDK paper.12345
At a glance
- Current roles
- Co-founder and Chief AI Officer +1 more
- Sector
- Company
- Fields
- Agents, Language models
- Education
- UIUC
- Work listed
- 2
- Sources cited
- 5
- Profile reviewed
Career
Wang previously interned at Tencent (2020, distributed deep learning), ByteDance (2021, the lightseq library), Microsoft (2022, unit-test generation) and Google Research (2023, multimodal LLM pre-training).1
Wang co-founded All Hands AI to build OpenHands and serves as its Chief AI Officer, while completing a PhD at UIUC with Heng Ji. Wang's stated interest is AI that improves itself from real-world use, including turning user interactions into supervision for models.12
Research
Wang's CodeAct paper (2024) proposed executable code as a unified action space for LLM agents. Wang was then first author of the 2024 OpenHands paper, describing an open platform for AI software developers, and of the 2025 paper on the OpenHands Software Agent SDK.435
Work and contributions
OpenHands Open-source code
First author of the OpenHands paper (2024) and of the OpenHands Software Agent SDK paper (2025), and a co-founder of the company that builds the project.351
CodeAct Published research
First author of "Executable Code Actions Elicit Better LLM Agents" (2024), which proposes using executable Python code as the action space for LLM agents.4
Timeline
- Interned at Tencent, working on distributed deep learning.1
- Interned at ByteDance, working on the lightseq library.1
- Interned at Microsoft, working on unit-test generation.1
- Interned at Google Research, working on multimodal LLM pre-training.1
- Posted "Executable Code Actions Elicit Better LLM Agents" (CodeAct) as first author.4
- Posted the OpenHands paper as first author.3
- Posted the OpenHands Software Agent SDK paper as first author.5
Other documented roles
Roles the sources document that are not confirmed as current.
- Intern working on multimodal LLM pre-training (2023), Google Research1
- Intern working on unit-test generation (2022), Microsoft1
Where they work
- All Hands AI 3 people in the wiki
- University of Illinois Urbana-Champaign 3 people in the wiki
Related people
- Graham NeubigAlso at All Hands AI
- Robert BrennanAlso at All Hands AI
- Jacob AndreouAgents · Language models
- TekniumLanguage models · Agents
- Hyung Won ChungLanguage models · Agents
- Jason WestonLanguage models · Agents
- Gabe PereyraAgents · Language models
- Leandro von WerraLanguage models · Agents
- Junyang LinLanguage models · Agents
- Noam BrownLanguage models · Agents
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
5 sources · profile reviewed Oct 22, 2026 · data updated Oct 22, 2026
- Xingyao Wang — homepage
- OpenHands — About (team)
- OpenHands: An Open Platform for AI Software Developers as Generalist Agents (arXiv 2407.16741)
- Executable Code Actions Elicit Better LLM Agents (arXiv 2402.01030)
- The OpenHands Software Agent SDK: A Composable and Extensible Foundation for Production Agents (arXiv 2511.03690)