Run a swarm of experiment agents on one hard question: AI agent prompt

Split a research or optimization question into many experiments, let agents run them in parallel for hours, and have them design follow-up experiments from the results. Third Moment Research runs 10,000 agents at once on 36 to 48 hour investigations with OpenAI's Codex, and one run cut a backtest from about three hours to a few minutes.

Curated by AgentPlaybook Real Estate · How we make and check prompts

ResearchParallel experimentsadvanced

The prompt

You are the lead researcher coordinating a batch of experiment agents. Question: [e.g. why is our backtest slow / which pricing page converts best]. Data and code you may use: [describe]. Budget: [number of parallel runs, time limit]. Step 1: list 5-10 hypotheses, each with a single experiment that could confirm or kill it and the metric to measure. Step 2: write a self-contained brief for each experiment agent (inputs, exact steps, what to report). Step 3: after I paste back results, rank the hypotheses, explain what the evidence shows, and propose the next round of follow-up experiments. Keep a running log of what was tried and ruled out so no agent repeats work.

Open in ChatGPT Open in Claude Perplexity Download as agent skill (SKILL.md)

How to use it

Copy the prompt above.
Replace every [bracket] with your own details (or fill in the blanks on AgentPlaybook Real Estate, which remembers your answers for the next prompt).
Paste it into Codex, ChatGPT, Claude Code and review the answer before you use it.

Works with: Codex, ChatGPT, Claude Code

Real-world example: openai.com/index/third-moment-research/

Before you use it: what to check

Check yourself

Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically".
Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details.

Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.

</> Embed this prompt on your site

Paste this HTML into your blog or website. It's a plain card with no tracking scripts that links back to this prompt.

<blockquote style="border:1px solid #e4e4ee;border-left:4px solid #5b3fd9;border-radius:10px;padding:14px 16px;margin:16px 0;font-family:system-ui,sans-serif;max-width:640px"><p style="margin:0 0 6px;font-weight:700">Run a swarm of experiment agents on one hard question</p><p style="margin:0 0 10px;color:#5c6078;font-size:14px">Split a research or optimization question into many experiments, let agents run them in parallel for hours, and have them design follow-up experiments from the results. Third Moment Research runs 10,000 agents at once on 36 to 48 hour investigations with OpenAI&#x27;s Codex, and one run cut a backtest from about three hours to a few minutes.</p><a href="https://agentic-library.netlify.app/prompt/al-174-run-a-swarm-of-experiment-agents-on-one-hard-question/?utm_source=embed&amp;utm_medium=referral&amp;utm_campaign=prompt_embed" style="color:#5b3fd9;font-weight:600">Get the free AI prompt on AgentPlaybook Real Estate →</a></blockquote>

Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.

Get the free Solo Founder Stack pack

32 copy-paste agent prompts (PDF) plus 5 new prompts every week.

Get it free

More Research prompts

Market size & trend reportMarket research · beginner Compare software toolsTool comparison · beginner Interview insights synthesizerCustomer research · intermediate Stress-test a technical document before you share itDocument verification · intermediate

See all Research prompts →