My AGENTIC LOOP

Company -like Operating System for running a wide array of agentic LLMs, with me as the human in the loop.
AI Company OS — full architecture
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B

Most AI setups are a collection of tools that happen to be in the same browser. I try to create a system and expand it: advisors that hold their domains and share what they decide, an executive layer that turns decisions into shipped work, an automation layer that moves work between them, and an evaluation layer that scores everything and proposes its own improvements.

It runs on cloud tiers and on some hardware I already owned for cost efficiency on certain tasks. It was built inside-out over six layers. 

This setup is continuously expanding. More to add in the upcoming weeks. Currently working on my personal and entrepreneural GraphRAG

Layer

What runs there

What it's for

Local

Llama 3.2 (3B), Qwen 3.5 (4B) via Ollama — a spare laptop, 6GB GPU

A Free Play setup for various local LLM tests; LLMs can be replaced with others or can be extended with a different setup. Used mostly for Triage, classification, routing, reformatting, extraction. Simple and recurring tasks at zero marginal cost.

Free-tier cloud

NVIDIA NIM behind a gateway

Occasional use if needed as NIM is for development tests only. First-pass drafting, structured extraction, and the judge.

Premium

Claude, Gemini

Core reasoning, the coding agents where one wrong edit cascades, and the advisory layer. Load balanced.

Execution

Claude Code

The executive team — orchestrator, heads, sub-agents, all with scoped tool permissions.

Automation

n8n

Watching folders, moving work between the two halves, scheduled runs.

Sync layer

Google Drive + Drive for Desktop

The shared state. Reachable by Claude, by Gemini, and as a plain local path by anything running on my machine.
The brain team. AI advisors with real functional identities — technology, product, marketing, finance, venture strategy. These are the ones I brainstorm with rather than delegate to. They hold context across sessions, they push back, and they coordinate through a shared folder rather than an integration.
AI Company OS — phase 1 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
AI Company OS — phase 4 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
The execution team. Where a decision becomes work. An orchestrator, functional heads beneath it, and sub-agents beneath them. Safety here is enforced through tool permissions rather than instructions: research agents are read-only, only engineering agents get shell access, only heads can delegate.
The nervous system. The automation layer plus a shared file-based memory that both halves read and write. Work crosses between thinking and doing without a human carrying it.
AI Company OS — phase 5 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
AI Company OS — phase 6 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
Evals and observability. Traces on every run, and a judge model scoring output against rubrics calibrated against real outputs. Regression checks sit here too, so improving one agent doesn't quietly break another.
The self-improvement loop. Eval findings become proposals — a prompt edit, a rubric change, a model-tier promotion — queued for approval. Approved changes are applied, and that agent is permanently better. Fifteen to thirty minutes of review in a morning.
AI Company OS — phase 7 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
AI Company OS — phase 8 diagram
Me — Leader Human in the LoopBrain team (C-level Advisory)Shared memoryHead of Claude CodeProductEngMktgspecprotoresearchcodertestsreviewcampaignQueueSelf-improveprops →gateEvalsTracesJudgeFailanalyzen8nCronMCPHooksExternAmpli-tudeNIMCI srcOutput — code · docs · apps ROSTER CLOUD · PAID Claude Gemini FREE TIER · CAPABLE NVIDIA NIM LOCAL · FREE PLAY Llama 3.2 · 3B Qwen 3.5 · 4B
External reach. Analytics, ad platforms, video and email APIs. Last layer in the whole system. Expands or shrinks from time to time due to active needs.