I kept seeing the word “agent” everywhere in agentic AI. Which got me thinking — can I have more than one?

Turns out I can. After chatting with Claude and doing some Googling, I found a pattern called a research swarm: multiple agents, each assigned a different task, all working in parallel. The idea immediately clicked. What are we waiting for? I wanted agents. Many more agents.

Just to be conservative — and not kill my 16GB Mac Mini — I started with five:

Meet My Agents

AIResearcher

Monitors AI and technology news overnight. Its job is narrow: surface anything in the AI/tech space that has market relevance — model releases, regulatory moves, big enterprise deals, significant research papers. It feeds its findings to OvernightAnalyst.

TechResearcher

Goes deeper on specific technical stories. Where AIResearcher casts a wide net, TechResearcher dives into the details — a zero-day vulnerability and its blast radius, a product launch and its competitive implications, a policy change and who it affects. Also feeds OvernightAnalyst.

FinanceCrawler

Covers the pure financial side: pre-market futures, earnings releases, Fed statements, bond yields, commodity prices, and major international market closes. Runs in parallel with the tech agents, feeds into OvernightAnalyst.

OvernightAnalyst

The synthesizer. Receives outputs from AIResearcher, TechResearcher, and FinanceCrawler and merges them into a coherent picture: what matters today, what connects to what, and what the combined signal means for the market opening. Does not generate stock picks — that’s StockPicker’s job.

StockPicker

The final output layer. Reads OvernightAnalyst’s synthesis and generates 3–5 specific trade ideas with a paragraph of rationale for each. Picks are grounded in the overnight research — not random recommendations, but specific thesis-driven ideas tied to what actually happened while the market was closed.

These aren’t chatbots you have to prompt. They’re persistent agents that wake up on a schedule, do their job, hand off to the next agent in the chain, and send the final output to me — automatically.

Why separate roles?
A single agent trying to do all of this would produce mediocre output at each step — too broad to go deep, too busy to synthesize well. Specialization lets each agent build domain expertise over time. StockPicker gets sharper because it only has to be good at one thing: reading a synthesis and generating picks.

The Pipeline

The message flow runs like this: FinanceCrawler and TechResearcher work in parallel overnight, both feeding into OvernightAnalyst. OvernightAnalyst synthesizes and sends to StockPicker. StockPicker sends me the morning brief.

It's a hand-off chain. Each agent has a specific job and passes its output downstream. No one agent has to do everything, so each can do its part well.

What surprised me
Each agent builds its own memory over time. After a week, StockPicker had learned which sectors I was most interested in, which news sources tended to produce actionable signals, and which types of picks I'd followed up on. The briefings got noticeably sharper.

What a Morning Brief Actually Looks Like

A typical 6 AM delivery includes: overnight macro developments (Fed speakers, international market closes, commodity moves), 2-3 tech/AI stories with market implications, and 3-5 stock picks with a one-paragraph rationale for each. The whole thing takes about two minutes to read.

Is it perfect? No. But it's better than starting my morning cold with no context. And it's free — the research swarm runs on its own while I sleep.

What This Made Me Realize

The hard part of building this wasn't technical — it was knowing what I wanted. Once I could describe the workflow ("overnight research → synthesis → picks delivered before market open"), the implementation was straightforward. The bottleneck was articulating the goal clearly enough for an agent to execute it.

That turned out to be a theme across everything I built in the weeks that followed.