Research hundreds of people and build slide decks at scale
Guide describes using Manus for large research jobs like profiling the top 250 AI researchers, turning event exports into slide decks, and deploying websites.
What they did
The post says Manus handles big research jobs by splitting them into many parallel sub-tasks, like spinning up 250 sub-agents to profile one AI researcher each and finishing in 10–15 minutes. It also describes feeding in structured inputs such as a Luma event export or a Google Sheet, then having Manus scrape sites, call connected tools like SimilarWeb, and assemble outputs such as slide decks or a deployed searchable website; the person provides the prompt, files, templates, APIs, and any brand rules.
How it went
The reported outputs are production-ready assets at scale: a 250-profile research pass in 10–15 minutes, event-export slide decks with charts, and public websites built from a sheet of 100 papers. The post presents these as examples and prompt patterns, but gives little detail on verification or failures.
Worth knowing
The main catch is setup: results depend on supplying connectors, attached templates/files, and clear project or knowledge rules so Manus follows brand and formatting requirements.
Labels
Manus handles big research jobs by splitting them into many parallel sub-tasks, like spinning up 250 sub-agents to profile one AI researcher each
having Manus scrape sites, call connected tools like SimilarWeb, and assemble outputs such as slide decks or a deployed searchable website
The post presents these as examples and prompt patterns
The “Material” label failed its quality check and is not used for selection.
Details
- Agents
- Manus
- Source
- Source date
- Feb 16, 2026
- Accounts
- 1
- In registry since
- Oct 16, 2026