Core Contributor

The Core Contributor journey is for engineers who want to work directly on FEAGI internals, especially feagi-core architecture, neural runtime behavior, and platform portability. This section helps you move from first-time codebase orientation to productive development with the synced architecture corpus under feagi-core-source.

feagi-core Architecture at a Glance

From the architecture sources, FEAGI core is organized as layered Rust crates with clear boundaries between foundational data structures, runtime state, neural computation, and integration surfaces.

• Foundational types and shared data structures feed all higher layers
• State management provides deterministic runtime coordination
• Algorithm crates handle neurodevelopment, burst execution, and plasticity
• Application and I/O surfaces compose these crates for server operation
• Feature-gated builds support desktop, cloud, embedded, and WASM targets

Start with:

• Core architecture overview
• Final crate architecture

Development Environment Setup

Use the FEAGI monorepo and project-local toolchains for reproducible builds.

1. Clone FEAGI-2.0 and enter the feagi-core subproject
2. Install Rust toolchain and required targets via rustup
3. Create and activate the Python virtual environment where Python tooling is required
4. Install development dependencies from project metadata (not global installs)
5. Run formatting, linting, and tests before opening a pull request

Running Tests and Quality Checks

Run checks from the relevant subproject directories:

• cargo test --workspace for Rust unit and integration coverage
• cargo bench --workspace for performance-sensitive pathways
• ruff check and mypy where Python support code is touched
• Architecture and platform-compliance checks documented in the source docs

If a command changes over time, prefer the command specified in the synced source doc nearest to the component you are modifying.

Curated Index Into feagi-core-source

The synced folder contains many architecture records and implementation notes. Use this curated map to find the right document quickly.

Start Here

• ARCHITECTURE.md
• FINAL_RUST_CRATE_ARCHITECTURE.md
• API_DESIGN_ARCHITECTURE.md
• VERSIONING_QUICK_REF.md

NPU, Performance, and Runtime

• GPU_IMPLEMENTATION_STATUS.md
• GPU_SUPPORT_STATE_ANALYSIS.md
• GPU_BACKEND_INTEGRATION_NEXT_STEP.md
• UNIFIED_OBSERVABILITY_ARCHITECTURE.md
• TEST_STATUS_AND_NEXT_STEPS.md

Genome and Development Pipeline

• GENOME_SCHEMA_VERSIONING.md
• SPATIAL_GRADIENT_IMPLEMENTATION.md
• MEMORY_NEURON_VISUALIZATION_WORKFLOW.md
• CRITICAL_BUG_FIX_SYNAPTIC_CONTRIBUTION.md

Training and Evolution

• FEAGI_TRAINER_ARCHITECTURE_AND_DESIGN.md
• FEAGI_TRAINER_TRAINING_PARADIGMS.md
• FEAGI_TRAINER_ADR_SET.md
• EXPERIENCE_TRAINER_E2E_IMPLEMENTATION_PLAN.md

Architecture Decisions and Change Records

• ARCHITECTURE_DECISION_INT8_DEFAULT.md
• COMPREHENSIVE_RUST_MIGRATION_PLAN.md
• INDEPENDENT_VERSIONING.md
• CRATE_GRAPH_SNAPSHOT_PROCESS.md — includes instructions for generating crate graph snapshots; versioned JSON snapshots live under feagi-core-source/crate-graphs/

Additional Implementation Notes

• LOGGING_STANDARDS_ENFORCEMENT.md
• GENERIC_INTEGRATION_COMPLETE.md
• INTEGRATION_TEST_RESULTS_FINAL.md
• DEBUGGING_INFRASTRUCTURE_PROPOSAL.md
• BUG_FIXES_2025_10_30.md

How To Use This Section Effectively

• Use this page as your starting map
• Open one architecture overview document before touching code
• Narrow to the subsystem-specific records listed above
• Keep source and docs aligned when your changes alter behavior or architecture