I build local and hybrid AI systems, bounded agent workflows, and source-first knowledge architectures.
This profile repository is the public entry surface for the ecosystem around Tree of Sophia (ToS), Agents of Abyss (AoA), abyss-stack, ATM10-Agent, and the public AoA layer federation. It helps humans and agents find the repositories that own the real charters, roadmaps, workflows, and implementations. It is a coordination surface, not the source of truth for those layers.
Two long-horizon directions anchor most of the work:
- Tree of Sophia (ToS): a source-first living knowledge architecture for philosophy and world thought
- Agents of Abyss (AoA): an operational federation of explicit layers for long-horizon agentic systems
Beneath them sits abyss-stack, the runtime substrate. Alongside them sit ATM10-Agent, a local-first companion surface, and aoa-sdk, a typed local-first consumer spine for source-owned AoA surfaces.
For a compact ecosystem vocabulary, see GLOSSARY.md.
- for the knowledge world and source-first architecture of thought, start with Tree-of-Sophia
- for the ecosystem center, constitutional map, and federation rules, start with Agents-of-Abyss
- for the runtime body beneath AoA and ToS, go to abyss-stack
- for the local-first companion application surface, go to ATM10-Agent
- for typed local-first federation reads and orchestration helpers, go to aoa-sdk
- for reusable practice, execution, and proof, move through aoa-techniques -> aoa-skills -> aoa-evals
-
Tree-of-Sophia
Living knowledge architecture for philosophy and world thought -
Agents-of-Abyss
Constitutional center of AoA: ecosystem identity, layer map, federation rules, and program-level direction -
abyss-stack
Infrastructure substrate for AoA and ToS: runtime, deployment, storage, lifecycle, and service posture -
ATM10-Agent
Local-first ATM10 companion: perception, memory, safe automation, voice, and operator surfaces
-
aoa-techniques
Reusable engineering practice -
aoa-skills
Bounded agent-facing execution workflows -
aoa-evals
Portable proof surfaces for bounded claims -
aoa-routing
Thin routing and dispatch layer across AoA surfaces -
aoa-memo
Reviewable, provenance-aware memory and recall surfaces -
aoa-agents
Role contracts, agent posture, and handoff boundaries -
aoa-playbooks
Scenario composition, recurring operations, and method surfaces -
aoa-kag
Provenance-aware derived knowledge substrate and retrieval-ready structures
- aoa-sdk
Typed Python SDK and local-first consumer spine for source-owned AoA surfaces
I am building toward:
- agent systems that stay legible as they scale
- knowledge architectures that accumulate layers without collapsing into noise
- local-first and hybrid systems that keep human meaning and operational clarity in view
- keep source-of-truth boundaries explicit
- publish durable techniques and reviewable workflows
- prefer modular growth over brittle fusion
- keep systems legible to humans while accelerating agents
- Systems: Fedora, Windows 11, WSL2, rootless Podman
- Languages: Python, Bash, JavaScript, PowerShell
- App and orchestration: FastAPI, Streamlit, LiteLLM, LangChain, LangGraph, n8n
- Inference and serving: Ollama, llama.cpp, OpenVINO / OVMS
- Data and memory: Postgres, Redis, Neo4j, Qdrant
- Observability: Grafana, Prometheus, Alertmanager
- Build workflow: ChatGPT, Codex, GitHub


