Google Cloud recently featured Vionlabs founder Arash Pendari in a video exploring how we use Vertex AI and Gemini to power content intelligence for streaming platforms worldwide. Watch the full video here.
Below is a summary of what we discussed and why this partnership matters for the media and entertainment industry.
The Discovery Problem in Streaming
Streaming platforms invest heavily in content. But when viewers can only find a fraction of what's available, that investment doesn't pay off. One of the main reasons people leave a streaming service today is simple: they can't find what they want to watch.
Traditional sorting relies on broad genre tags that are subjective and manually applied. An algorithm might see that a viewer watched three "action" movies, but those films could have completely different pacing, visual styles, and emotional journeys. Genre alone can't capture that.
The biggest streaming services have massive internal teams dedicated to solving this problem. Most platforms don't have that luxury. That's the gap Vionlabs was built to close.
"Companies invest a lot of money into content, and when your users are only seeing 20% of your library, that means 80% of your content isn't being viewed. We use Gemini to help viewers find what they want and streaming services invest in content their viewers want."
Arash Pendari, Founder & Creative Director, Vionlabs
What Vionlabs Does
Our core mission is to index content libraries with useful, structured data so that anyone in an organization can get value from their catalog. We use AI to break down video content scene by scene, extracting mood, emotion, pacing, characters, and narrative structure to build what we call an emotional fingerprint of every title.
But not every data point matters. It's about tagging the signals that are important for the content, not tagging everything. The goal is to give more context and more understanding to power better recommendations, smarter editorial curation, and more engaging discovery experiences.
The results speak for themselves. Our clients have run A/B tests showing increased content completion rates, meaning viewers are happier with what they're being recommended and are watching more of it. They're also seeing improved retention. It all comes back to the user experience: the better the experience, the better the numbers.
Why We Chose Google Cloud
From the start, we've been deliberate about what we build in-house and where we leverage external infrastructure. We build our own AI models because understanding content is our domain expertise. But we use Google Cloud's Vertex AI framework to train all of those models.
"We built our own AI models, but we've used Vertex AI's framework to train them. This means we can fine-tune our models quickly, review the results, and adjust fast. As a startup, that's really crucial."
Arash Pendari
This flexibility has been essential for working across different content types and regions. When we needed to fine-tune our models for anime content, for example, Vertex AI gave us the ability to do that quickly and see results fast. We're content experts, and Google Cloud lets us focus our energy where it matters most.
We also use Gemini for multi-modal processing and BigQuery as our data backbone. Combined with Vertex AI's training infrastructure, these tools have saved us a significant amount of time and investment that we would have otherwise spent building and maintaining our own AI framework.
More Than a Technical Partner
Google Cloud hasn't just been a technology provider for us. They've also been a genuine business partner, helping us scale into new regions where we don't have a direct sales presence. That kind of support is hard to find and incredibly valuable for a company at our stage.
"Google Cloud has been both a technical partner and a business partner, helping us scale to reach new clientele and new regions."
Arash Pendari
Vionlabs is also available on the Google Cloud Marketplace, making it easy for media companies to procure and deploy our AI directly within their existing Google Cloud environment.
What This Means for Streaming Platforms
The larger streaming services have dedicated teams and proprietary systems for content intelligence. Smaller and mid-sized platforms typically don't. That's where Vionlabs comes in.
With Google Cloud's infrastructure behind us, we can offer any streaming service the same level of content understanding that the biggest players have built internally. Scene-level AI that captures mood, emotion, and narrative. Metadata that powers better recommendations. Thumbnails and preview clips that drive engagement. Ad break placement that respects the story.
As Arash put it in the video: the AI is there to curate, but humans are there to perfect. We scale the creativity so your team can focus on what they do best.
Watch the full Google Cloud x Vionlabs video: YouTube
Ready to see what Vionlabs can do for your content library? Book a demo to see it in action.
.png)
.png)
