AI Content Creator and Publisher
An engine that finds trends on X, writes an on-brand article, spins up a video and a LinkedIn post, and syndicates variants to third-party sites — all published automatically on a schedule.
4 channels
article, video, LinkedIn post and syndicated variants from one topic
Set & forget
the full multi-channel pipeline runs unattended on a schedule
On-brand
every asset written and produced in the brand's own voice
Vendor-neutral
every AI and data provider swappable behind capability names
The brief
B2B Technology Brand
Marketing & Content (SaaS)An engine that finds trends on X, writes an on-brand article, spins up a video and a LinkedIn post, and syndicates variants to third-party sites — all published automatically on a schedule.
The challenge
What needed to change
Publishing a steady stream of high-quality, on-brand thought leadership is one of the most reliable ways to build authority — and one of the hardest habits to keep. It means constantly spotting what's timely, researching an angle, writing something genuinely useful, sourcing artwork, formatting it for the blog and hitting publish, week after week. For a busy B2B team that cadence is the first thing to slip. Our client wanted the output of a full content team without the manual grind: relevant, well-written articles appearing on their blog automatically, in their own voice, and reliably enough to run unattended — while never reading as generic AI filler.
Key requirements
What we needed to deliver
- Discover timely, high-interest topics by monitoring trending conversations on X.
- Generate a full, well-structured article in the brand's own voice — not thin, generic AI copy, with matching imagery sourced or generated automatically.
- Turn each article into a video, with a choice of formats — from an avatar presenter to a composed, studio-style video — complete with AI voiceover.
- Create and publish a tailored LinkedIn post for every piece.
- Produce article variants and syndicate them to third-party publishing sites to widen reach.
- Publish every asset — article, video, social post — straight to the brand's own channels with no manual step.
- Run the entire multi-channel pipeline on a schedule for full, hands-off automation, as well as on demand.
- Handle long-running article and video generation reliably, without timeouts or silent failures.
- Support multiple organisations and projects, each with its own connected destinations, settings and subscription tier.
- Keep every underlying AI and data provider behind neutral capability names, so vendors can be swapped without affecting users.
The solution
How we solved it
iCentric built an automated, multi-channel content engine that takes a single trending topic and turns it into a whole campaign — an article, a video, a social post and syndicated variants — publishing every piece to the right place without a human in the loop. Point it at the subjects a brand cares about, set a schedule, and it runs itself.
It starts with what's trending on X
Each run begins by monitoring X for high-engagement, recent conversation on every subject the brand has chosen to cover. Large language models read that signal, pick the most timely angle, and write a full, structured article — around 900 words — in the brand's established voice, ready for the blog. A fitting hero image is sourced from a stock library or generated by AI to go alongside it.
One topic becomes a whole campaign
From that single article the engine fans out across every channel automatically. It publishes the finished article, correctly formatted, straight to the brand's website. It turns the piece into a video — with a choice of formats, from an avatar presenter to a composed, studio-style video, complete with AI voiceover — and publishes that to the website and the brand's video channel. It writes a tailored LinkedIn post and publishes it to the brand's LinkedIn presence. And it produces article variants and syndicates them to third-party publishing sites to widen reach. One trending topic becomes a coordinated set of assets across web, video and social — every one of them on-brand.
Set the schedule and it runs itself
The whole pipeline is built for full automation. A brand chooses which subjects to watch, which channels to publish to and how often to run — and the scheduler does the rest, discovering trends, generating every asset and publishing across all channels on a recurring cadence, unattended. Runs can also be triggered on demand at any time. Because writing articles and producing video takes real time, the heavy work runs on a dedicated long-running worker with a durable database, so scheduled runs complete reliably instead of timing out, and the system remembers its settings, history and the assets it has already produced.
A multi-tenant product, vendor-neutral by design
What began as one brand's pipeline was generalised into a multi-tenant platform: separate organisations and projects, each with its own connected destinations, settings and subscription tier. A core design rule is that no underlying provider is ever exposed to the user — every external service, from the AI writer to the voiceover and video engines, sits behind an iCentric-defined capability name, so any provider can be changed at any time without users noticing and without the product having to change.
The outcome
The impact
The brand now runs an entire multi-channel content operation on autopilot. From a single trending topic it automatically publishes an article and a video to its website, a post to LinkedIn, and syndicated variants to third-party sites — on a schedule, in its own voice, with no day-to-day effort. What used to need a writer, a designer, a video editor and a social manager now runs as one unattended pipeline. And because the whole thing was built as a vendor-neutral, multi-tenant platform, iCentric can run this same content machine for many organisations at once — each with its own channels, cadence and brand voice.
Get in touch today
Book a call at a time to suit you, or fill out our enquiry form or get in touch using the contact details below