ReelRecon: competitor reels turned into searchable notes
Downloads a profile's reels, transcribes them and tags the hooks and angles in a searchable dashboard.
- System and automation
- In-house R&D, prototype
- In-house prototype, open source
At a glance
Client ReelRecon, an OutGrow open-source project
Client
ReelRecon, an OutGrow open-source project
Industry Social media research
Industry
Social media research
Year 2025
Year
2025
Status In-house prototype, open source
Status
In-house prototype, open source
Every case on this site answers the same four questions, so you can compare them.
Before
Teams track competitors by watching reels and copying scripts by hand. Nothing ends up searchable.
What it does now
The tool downloads a profile's reels.
The audio is extracted and transcribed with Whisper.
An LLM tags each reel's hook, angle and format.
Everything lands in a dashboard with full-text search and CSV export.
Where a person decides
A person reads the tagged reels and decides what is worth acting on. The tool only collects and sorts.
Result
In-house prototype. No outcome figures yet.
How we built it
Wrote the downloaders in Python.
Used FFmpeg for audio and Whisper for transcripts.
Wrote prompts that return structured JSON for each reel.
Built the React dashboard and scheduled weekly crawls.
Stack
- Python
- FFmpeg
- Whisper
- Groq
- React
- PostgreSQL
From the project
More work like this
One inbox for every inquiry a Dubai studio gets
An email engine that pauses itself before it hurts the domain
A distributor's back office rebuilt to sell nationally
Your case study starts with an audit.
One to two weeks inside your operations. You get a map of where time and money leak, the return on each fix, and a fixed scope before we build anything.
Most businesses we meet have paid for an AI tool nobody uses. We start from your numbers, so the first build is chosen by payback.
Zain Khan and Anwar Sheikh
Founders, OutGrow
Outgrow
Intelligence
Studios.