Prove it in n8n, then productise
Each automation started as an n8n workflow; once stable, its exact settings were moved into a dedicated console with editing and scheduling.
Pyalm HQ: AI-written, auto-published social content from live news
An automation console that turns filtered industry news into AI-written LinkedIn and Instagram posts and publishes them on a schedule.
Global
The challenge
What needed solving.
Consistent, relevant social posting takes hours each week. We wanted daily posts tailored to different audiences without a content team — and without repeating the same story twice.
What we built- Scheduled runs that read RSS feeds and search queries
- Age and keyword filters plus a 30-day ledger to drop already-covered stories
- AI step that writes each post for a defined audience and channel
- Publishing to LinkedIn (company and personal) and Instagram through Tenreply
- Run logs and Telegram notifications for every execution
- Each automation first proven as an n8n workflow, then moved into the console
Deduplication ledger
Recording every published story for 30 days stops the AI from posting about the same news twice, the most common failure in automated content.
The result
Pyalm publishes audience-specific posts daily with no manual writing, and every run is logged and auditable.
Related work
Similar projects.
Wedding photography platform
Photographers upload thousands of event photos; guests scan a QR code, take a selfie, and instantly see every photo they appear in.
Read the case studyRetail property collections team
A stakeholder-ready product prototype that ranks rent decisions by tenant sales and affordability, with evidence, approvals, and an audit trail.
Read the case study