Huang Ruiteng
Always-on agent teams, governed by human judgment. LoopX is a dynamic goal control plane for long-running AI agents that keeps objectives, gates, and evidence stable across turns.
At a Glance
- AI researchers
- Open-source maintainers
- Software engineers
- Solo creators/operators
AI Tools by Huang Ruiteng
(1)
LoopX
AI Agent Loop Control Plane
Discussions
No discussions yet
Be the first to start a discussion about Huang Ruiteng
Latest News
Support for Pi agent as a first-class LoopX host integration
Hosted documentation portal launched
LoopX public homepage launched
LoopX v0.2.0 control plane ships
Products & Services
A lightweight state kernel and agent-agnostic local control plane for loop engineering, designed to keep long-running AI agent work reviewable and manageable.
Market Position
A 'lightweight state kernel' that focuses on the control plane (goals/gates) rather than agent execution, positioning it as an agnostic layer above runtimes like Claude Code or Codex.
Leadership
Founders
Huang Ruiteng
Senior Machine Learning Engineer for AML at ByteDance, core contributor to OpenViking. Previously associated with Tsinghua University.
Executive Team
Huang Ruiteng
Creator
Senior Machine Learning Engineer at ByteDance and core contributor to OpenViking.
Founding Story
Started in May 2026 as a scaffold to anchor long-running agent work in explicit state. It was renamed to LoopX in June 2026 and evolved to solve the 'hard control-plane problems' of AI agent autonomy and human oversight.
Business Model
Revenue Model
Open-source (MIT License)
Target Markets
- AI researchers
- Open-source maintainers
- Software engineers
- Solo creators/operators
- Multi-day engineering, research, or experiment loops
- Long-running issue and PR resolution
- Recurring heartbeat or monitor tasks
- Coordinating peer-agent teams
- Creator and office operations workflows
- OpenViking contributors