Hive
Built the client for Aden's open-source agentic framework. Goal-driven AI agents that self-improve — define outcomes in natural language, auto-generate agent graphs, and evolve on failure.
Next.jsTypeScriptReact FlowWebSocketPythonTailwind CSS
Visit Live Project
About the project
Hive is an outcome-driven agent development framework by Aden (YC-backed). Instead of hardcoding workflows, you describe goals in natural language and the framework auto-generates agent graphs with dynamically created connection code. When agents fail, the system captures failure data, evolves the graph, and redeploys automatically.
Built the client application — the interactive dashboard, TUI, real-time WebSocket monitoring, and the visual agent graph interface. The framework supports 100+ LLM providers via LiteLLM, human-in-the-loop intervention nodes, granular cost controls, and SDK-wrapped nodes with shared memory, monitoring, and tool access out of the box.
Story
Short-term contract for Aden, a YC-backed startup building an open-source agentic framework. They had the backend and the framework architecture figured out but needed someone to ship the client fast.
I built the interactive dashboard — real-time WebSocket streaming that shows agent execution as it happens, and a visual graph editor using React Flow where you can see how agents connect, where failures occur, and how the system evolves itself. The framework is outcome-driven: describe goals in natural language, it auto-generates agent graphs, and when something fails it captures the data and redeploys an evolved version.
Shipped under tight timelines. The client now powers the open-source Hive framework on GitHub under Apache 2.0. It was a good engagement — fast-paced, clear scope, and the team knew exactly what they needed.
Screenshots
Enlarge
Demo Video
Technology Stack
Next.jsTypeScriptReact FlowWebSocketPythonTailwind CSS
Interested in this project?
Check out the live demo or get in touch to learn more.