Building the Agentic Science Era Together
Join a global community of researchers, developers, and scientists bridging the gap between AI agents and petabyte-scale data.
"We believe that context is the currency of intelligence. In a world drowning in data but starved for meaning, we are building the infrastructure that lets agents reason rather than just retrieve. This isn't just about faster I/O—it's about enabling a new kind of science where agents are first-class citizens."
Why Contribute
Collaborative Innovation
Work alongside climate scientists, AI researchers, systems engineers, and domain experts pushing the boundaries of autonomous discovery.
Real-World Impact
Power agents that predict weather, discover materials, analyze cosmic events, and accelerate scientific breakthroughs across disciplines.
Open Standards
MCP standardizes how agents talk to tools and data. Contribute to the foundation of agentic computing infrastructure.
Cutting-Edge Research
Context engineering, hierarchical memory, agent orchestration—work on the frontier of AI and systems research.
Platform Status
Operational
99.99% Avail
High Latency
latest
Platform Evolution
Project Inception
NSF award ($5M), initial CLIO Ingest design
Prototype v0.1
First context transfer demo
IOWarp Core Beta
Public release, HDF5/NetCDF integration
Agentic Scale
Target: 10,000 concurrent agentsVision
Key Repositories
Community Spotlight
Climate Agent v2.0
500TB CMIP6 analysis with autonomous pattern discovery
HDF5-MCP
Direct HDF5 access for Claude agents
Context-Aware IO
Published SC'25
IOWarp Academy
Free course on agentic science infrastructure
Upcoming Events
Weekly Community Call
Feb 1210:00 AM PT
Zoom
Webinar: Agentic Science 101
Feb 2411:00 AM PT
Livestream
IOWarp Summit 2026
Mar 15All Day
Chicago
Contribution Pathways
Platform Development
Build core engines, optimize for CXL/RDMA, enhance context orchestration
Learn more →Scientific Demos
Agent workflows for climate, genomics, materials, and other domains
View demos →MCP Integrations
Connectors for tools, schedulers, formats, and scientific instruments
GitHub →Latest Releases
IOWarp Core: Performance improvements and bug fixes
IOWarp Platform: Docker deployment improvements
IOWarp Core: Initial public release
Open Governance
IOWarp follows an open governance model with Working Groups, a Technical Steering Committee, and a clear path from contributor to maintainer. All decisions are made transparently with community input.
Community Stats
Get Involved
Ready to get involved?
Join us in building the infrastructure for autonomous scientific discovery. Every contribution matters.