Varun Vastralaya Automation
Phone-first workflow that turns product photos into prepared social media posts.
- Type
- Automation
- Client
- Varun Vastralaya
- Role
- Workflow design · Automation engineering
- Year
- 2025
Evidence / status
Internal tool. Reported outcomes await verification.
Challenge
Content production without a content team
The workflow had to begin with a phone photo, remain understandable to a non-technical operator, and avoid a conventional video-editing process.
Constraints
- Input needed to be phone-first.
- The operator could not manage a technical dashboard.
- Captions needed to fit a regional Hinglish-speaking audience.
- Running costs needed to remain predictable.
Approach
Key decisions
Telegram as the control surface
A bot provides a familiar single-action input without requiring a custom application.
Still-image video generation
Product photos become short motion assets without a manual editing timeline.
Local caption generation
A local model prepares Hinglish copy without adding a per-request writing API cost.
Image preparation before generation
Input photos are normalised before later workflow stages to reduce inconsistent output.
Evidence
What can be stated
Current status: Internal tool. Unverified outcomes are labelled instead of presented as proof.
Posting-frequency, cost, and processing-time claims require owner verification.
Needs verification
Technology stack
- n8n
- Kling AI
- fal.ai
- Ollama
Next project
Continue through the work archive.
Start a project
Let's build something great
Drop your details below. We'll have it open in Gmail, ready to send — no copy-pasting.