Social
Build a targeted influencer list
Describe a creator segment and a Nimble Web Search Agent returns matching influencers with their handle, follower count, engagement, and profile URL. Running several queries accumulates a deduplicated dataset in Supabase that a dashboard filters by platform, category, and follower size, with CSV export for outreach.
Quick Start
Inputs
- Query A creator segment: niche plus platform plus follower band plus geography, for example "home-cooking creators on TikTok, 50k-500k followers, US".
Outputs
- A deduplicated influencer dataset in Supabase
- Handle, platform, and profile URL per creator
- Standardized follower counts and engagement
- A clean query-derived category per creator
- Public contact email when the creator lists one
- CSV export of any filtered view
How it works
A four-step discovery flow.
- Describe the segment Describe a creator segment: niche, platform, follower band, and geography, treated as hard filters.
- Discover the creators A dataset_building agent searches the named platforms and creator directories and returns matching creators with follower counts and profile URLs.
- Accumulate and dedupe Results are standardized and upserted into a Supabase table, deduplicated on platform and handle; more queries grow the dataset.
- Filter and export A dashboard filters by platform, category, and follower size and exports the current view to CSV.
Stack
Nimble primitives plus the full runtime stack.
Nimble APIs
What it does
- Web Search Agents One dataset_building agent per query discovers creators matching the niche, platform, follower band, and geography, with follower counts and profile URLs.
3rd Party Tools
Role
- Supabase The influencers table where the dataset accumulates across queries, deduplicated.
- Streamlit The dashboard: filter by platform, category, and follower size, then export CSV.
- pandas Dedup, follower parsing, and sorting.
- Python 3.10+ Resumable, concurrent discovery.
Reach out if you have any questions.
Talk to an Expert