Cody Menefee builds at Firecrawl and created OpenPasture, an open-source effort to support pasture-based livestock farming. His projects connect web-data collection and AI agents with practical decisions about grazing and farm equipment.[1][2][3]
Start with You’re Not Thinking Big Enough: Rebuilding Food Systems with AI Agents →
GitHub Background
In his conference presentation, Menefee describes moving from blue-collar work in Kentucky to becoming a self-taught engineer. His interest in pasture-based livestock farming led him to focus on a recurring practical problem: farmers must decide where animals should graze while taking account of grass recovery, weather and previous use. Moving a virtual fence is only part of that work; deciding where to place it requires information and judgment.[2]
Contributions
A grazing decision with reasons
OpenPasture combines open software with an initiative for virtual-fence collars. Its proposed farm workflow gathers environmental information, recommends whether a herd should stay or move, and sends a chosen boundary to collars. The project distinguishes its concept interface and roadmap from operating capabilities. It exposes software interfaces so another decision system can connect to the same equipment. This makes the boundary between observation, recommendation and physical action explicit.[3]
A request that becomes a data collector
Menefee describes building the initial Prometheus demo and then developing it with the Firecrawl team. A user asks for web data in ordinary language; the system returns collection code and sample data. Users can keep the code or have Firecrawl run and maintain the collector. The project connects an agent’s interpretation of a request to reproducible software that gathers the requested information.[4][5]
Tools, writing & talks
Writing, research & talks
Start here
A presentation on using software and agents to tackle the practical work of rotational grazing.
Watch the grazing problem and the distinction between moving a fence and choosing its position. No agriculture background is needed; it offers context before exploring the project’s code and hardware.
Speaker[2]- You’re Not Thinking Big Enough: Rebuilding Food Systems with AI Agents — Cody Menefee, Firecrawlyoutube.com
A Firecrawl engineer pitches replacing manual pasture inspection with a language model that reads animal locations, drought data, and camera imagery to recommend rotational grazing moves, built on a farming knowledge base scraped with Firecrawl, while naming real blockers like locked-down GPS collar
Catalog workOriginal source
Tools
- OpenPastureopenpasture.dev
Open software and collar hardware for pasture-based agriculture.
- Prometheusfirecrawl.dev
Turns web-data requests into collection code, with optional hosting and maintenance.