Pro Tips

How to Structure
Your AI Agents

JPMorgan published their multi-agent system. It looks almost exactly like the framework I use across my businesses. Here’s the rule, the 4-part architecture, and how to apply it as a solo operator.

The One Rule

Most beginners build one giant agent and try to make it do everything. It gets confused. It forgets things. It makes mistakes constantly.

The rule: one agent, one job. Like an actual employee with one specialty. Build a stack of those, then build one supervisor that ties them together.

JPMorgan calls this exact pattern a supervisor agent. I call mine my co-founder. Same architecture.

My Stack 5 Agents + 1 Co-Founder

For my e-commerce business, here’s the actual setup. Each agent only knows what it needs to do its one job.

Agent 01

Email Marketing Agent. Writes weekly emails, manages sequences, tracks opens/clicks.

Agent 02

Google Ads Agent. Bid management, keyword adds, ad copy variants, daily budget alerts.

Agent 03

Meta Ads Agent. Same job as #2 but for Facebook/Instagram. Different platform, different agent.

Agent 04

Customer Service Agent. Triages incoming questions, drafts replies, flags refunds for human review.

Agent 05

Daily Reporting Agent. Pulls data from all of the above into a single morning dashboard.

The Supervisor

Co-Founder Agent. Watches the others. Tells me when one breaks. Goes in and fixes it.

JPMorgan’s Pattern The 4-Part Architecture

JPMorgan’s multi-agent system — nicknamed Ask D.A.V.I.D. — was presented at LangChain Interrupt 2025. It has four core agent roles. They map directly to anything you’d build at home.

┌────────────────────┐ │ SUPERVISOR AGENT │ ←── you talk to this one │ (orchestrator) │ └─────────┬──────────┘ │ ┌─────────────┼─────────────┐ ↓ ↓ ↓ ┌──────────┐ ┌──────────┐ ┌──────────┐ │STRUCTURED│ │UNSTRUCTURED │ANALYTICS │ │ DATA │ │ DATA ││ AGENT │ │ │ │ (RAG) ││ │ │ SQL/APIs │ │ emails, ││ runs │ │ │ │ notes, ││ models / │ │ │ │ PDFs ││ code │ └──────────┘ └──────────┘ └──────────┘

Role 01

Supervisor Agent

The one you talk to. Understands intent, decides which sub-agent to call, holds short and long-term memory, escalates to a human when needed.

Role 02

Structured Data Agent

Translates natural language into SQL queries or API calls. Runs them. Summarizes the result. Use this for anything in a database or a SaaS API.

Role 03

Unstructured Data Agent (RAG)

Vectorizes your emails, meeting notes, PDFs, audio transcripts. Finds the right snippet. Returns it. This is the agent that makes “answer using my company’s docs” possible.

Role 04

Analytics Agent

Runs the actual computations — financial models, simulations, custom code. The supervisor calls it for anything that needs math, not just retrieval.

Solo Operator How To Apply It With No Engineering Team

You don’t need LangGraph or a 50-engineer team. Here’s the same pattern in a small-business stack.

Layer 01

Build with Claude Code

Way cheaper than enterprise agent platforms. You describe what each agent does, Claude builds it. One folder per agent.

Layer 02

Run them in OpenClaw

Hosts your agents and runs them on schedules. Each agent gets its own runtime, its own credentials, and its own logs. Cheaper than enterprise platforms; sturdier than running scripts on your laptop.

Layer 03

Talk to them through Telegram

One Telegram bot per agent (or one supervisor bot routing to all of them). I get my morning report, refund flags, ad alerts — all on my phone, no dashboard required.

Start Here Don’t Build All 5 Today

01

Pick the most painful job in your business

The one that bleeds your time every week. That’s your first agent. For most people it’s customer service or daily reporting.

02

Build that one agent. Only that one.

Spec it: input, output, schedule, what tools it can call. Build with Claude Code. Don’t add a second job to it.

03

Don’t add agent #2 until #1 actually works

Watch it run for a week. Fix what breaks. Then build the next one. The temptation is always to scale before you’ve stabilized. Resist.

Honest limitations

Multi-agent stacks need monitoring — if you don’t have a supervisor checking the others, errors compound silently. The supervisor itself needs guardrails (you don’t want it spending money or sending public messages without approval). And costs scale: 5 agents running daily is fine on Claude paid plans, but a supervisor that’s constantly checking everyone can hit token limits if you’re not careful with prompts.

For Your Job

Set Up Claude for Your Specific Job

If you’re ready to set up Claude for your specific job — with custom skills, connectors, and automations built around the work you do every day — I built a bootcamp just for you.

Start the Weekend Bootcamp →

Go Even Further

Join the AI Income Lab

If you’re looking to go even further, join mine and my husband’s community group where we give you all the AI agents and systems running our businesses.

Join the Community →

© 2026 Mariah Brunner. All rights reserved.