Ground Control / Research
The systems behind
reliable autonomy.
Ground Control develops compound AI systems for long-horizon execution: revision-pinned operational graphs, typed workflow compilation, hierarchical agent orchestration, policy-mediated effects, and feedback-guided harness optimization.
This collection specifies the system models, consistency contracts, execution protocols, and evaluation methods underlying that architecture, with explicit connections to distributed systems, programming languages, and agentic machine learning.
Explore the papersTechnical white papers
01—08 / Systems, execution, adaptation
Operational architecture
Revision-Pinned Operational Graphs for Long-Horizon Agents
Revision-addressed provenance, canonical content binding, and optimistic concurrency control
Implemented architecture · White paper 01
Read paper : Revision-Pinned Operational Graphs for Long-Horizon AgentsWork coordination
Transactional Plan Materialization
A controlled path from mission shaping to work packages
Implemented architecture · White paper 02
Read paper : Transactional Plan MaterializationWorkflow systems
Typed Compilation of Agent Workflows
Turning an authored graph into a checked, revisioned execution plan
Implemented architecture · White paper 03
Read paper : Typed Compilation of Agent WorkflowsEvaluation and governance
Subject-Bound Evaluation in Multi-Agent Systems
Evaluation provenance, configuration drift, and stale-evidence rejection across governed transitions
Implemented architecture · White paper 04
Read paper : Subject-Bound Evaluation in Multi-Agent SystemsAgent coordination
Resource-Aware Coordination of Persistent Agent Teams
Hierarchical multi-agent orchestration with dependency-aware scheduling and pressure-based admission control
Operating methodology · White paper 05
Read paper : Resource-Aware Coordination of Persistent Agent TeamsDurable execution
Durable Execution Across Agent Lifecycles
Revision-pinned configuration, lease-mediated attempt ownership, and crash-recovery semantics
Implemented architecture · White paper 06
Read paper : Durable Execution Across Agent LifecyclesPolicy and autonomy
Policy-Mediated Autonomy with Resumable Action Attempts
Admission control, deferred invocation, and replay-time authorization
Implemented architecture · White paper 07
Read paper : Policy-Mediated Autonomy with Resumable Action AttemptsSystem adaptation
Feedback-Guided Optimization of Agent Harnesses
Black-box system adaptation through persistent configuration and controlled evaluation
Operating methodology · White paper 08
Read paper : Feedback-Guided Optimization of Agent Harnesses