Systems Architecture May 2026 · Kika (Nica Loren)

How I Orchestrate My Entire Workflow

From Pieces Memory to Codex Agents to ChatGPT. I don't code alone anymore — I have a fleet.

May 2026 4 min read Published
The Agent Fleet
Pieces OS
Pieces OS
Long-term memory graph
Lapapi
Lapapi
Mission control hub
Codex
Codex
Builder agent
ChatGPT
ChatGPT
Strategic layer
Ollama
Ollama
Local inference
Memory Routing Building Strategy Local

Here's how it actually works — the full stack, the daily flow, and the philosophy that makes it all hold together.

Foundation

Pieces OS Is My Brain

Everything starts with Pieces OS. It's not just a snippet manager. It's my long-term memory graph. Every clipboard event, every screenshot OCR, every browser tab, every Codex session, every ChatGPT conversation — Pieces captures it, indexes it, and makes it searchable.

When I open a project I haven't touched in 3 weeks, I don't have to remember where I left off. I ask my system to pull a morning brief, and it synthesizes what I accomplished, open errors, active projects, and 2–3 next actions.

Routing

Lapapi — Personal Mission Control

Lapapi is the hub. A macOS-first dashboard with that unmistakable mustache — my signature icon for the app that connects Pieces, Ollama, Codex, and whatever model I need.

Every morning, Lapapi runs a health check: connector status, dirty git states, Downloads clutter, system memory pressure. If something's wrong, it tells me. If everything's green, it gets out of the way.

Build

Codex: The Builder Agent

Codex is where the real work happens. I recently removed Claude from my local-first stack entirely. Now: Pieces, Ollama, Apple local AI, and Codex project context only. Build passed. App relaunched. No regrets.

Fleet

Specialization Over Generalization

I don't use one AI for everything. I use agents — specialized personas:

AgentRole
HermesGateway / Telegram & Discord dispatch, cron jobs, infrastructure
KartieCode review, architecture decisions, system diagnostics
Neve-02Creative tasks, brand assets, Jez V marketing
CodexCode generation, project scaffolding, deployment
GooseLocal inference, Ollama model management, extensions
Strategy

ChatGPT: The Strategic Layer

ChatGPT is where I think. Codex is where I build. Pieces remembers everything.

Daily Flow

An Actual Day

  1. Morning brief (Lapapi + Pieces) → full context recovery
  2. Deep work (Codex) → reads AGENTS.md, implements next feature — zero prompting
  3. Creative sprint (ChatGPT + GPT Image 2) → brand assets, copy, positioning
  4. Evening wrap (Pieces memory capture) → the entire day's context is saved for tomorrow
Philosophy

Workflow Over Model

The model matters less than the workflow.

A mediocre model with perfect context routing beats a frontier model with zero memory. The model will change next month. Your workflow is what compounds.

If you're building with AI agents: invest in memory and routing before you invest in better models.