• conforms_to::[[Reference Form Contract]]
• serves_as::[[Runtime-Shaped Answer to the Compound AI Systems Framing]]
• in_domain::[[eOS Continuum]]
• authored_by::[[Christopher Allen]]
• has_lifecycle::[[Seed Stage]]
• has_curation::[[Working Draft]]

Towards Resource-Efficient Compound AI Systems (Chaudhry et al., 2025)

URL: https://arxiv.org/abs/2501.16634

The paper proposes a vision for resource-efficient compound AI systems through a declarative workflow programming model paired with an adaptive runtime system for dynamic scheduling and resource-aware decision-making. The contribution is the runtime framing: compound AI systems need a runtime that handles scheduling, resource allocation, and decision-making across components -- not just an orchestration layer over external services.

Adopted

The paper's runtime framing arrives close to this graph's substrate-vs-glue diagnostic. Compound AI systems integrating multiple components (LMs, retrievers, external tools) need a runtime that owns the scheduling and resource decisions; that runtime is in substrate territory, not orchestration territory. The eOS Continuum substrate goes further: the runtime owns not just scheduling and resource decisions but also state, atomicity, capability separation, multi-agent coherence, hot reload, sandboxed code load, asynchronous events, and direct state introspection.

Not adopted (yet)

The paper's runtime is concerned with scheduling and resource allocation; it does not name the eight runtime primitives this graph's [[Agent Runtimes Require Substrate Primitives, Not External Glue]] Conviction names. The runtime-LAYER framing is right; the specific substrate commitments (orthogonal persistence as foundational, single-coherence-domain by design, capability-OS-tradition lineage) are this project's contribution.

Sources

• arXiv: https://arxiv.org/abs/2501.16634
• Authors: Chaudhry et al.
• Date: 2025-01
• Venue: arXiv preprint
• Stub note: Authored 2026-05-03 as one node in the compound-AI-systems literature stub set. Body to be expanded with the declarative workflow programming model details and the per-substrate-primitive correspondence in a future session.

Relations

• conforms_to::[[Reference Form Contract]]

  • One node in the compound-AI-systems literature stub set.

• informs_downstream::[[Agent Runtimes Require Substrate Primitives, Not External Glue]]

  • The paper's runtime-shaped answer is one architectural layer above the substrate-LAYER answer eOS Continuum gives; both name the runtime as the right place for these concerns.

• composes_with::[[Compound AI Systems Optimization Survey (Lee et al., 2025)]]

  • Sibling in the compound-AI-systems literature; the survey covers methods like the one this paper proposes.