AI is everywhere right now. Everyone’s talking about it, experimenting with it, and trying to figure out how it actually fits into real workflows.
For us, the question was simple:
How do we use AI to create more content, faster—without sacrificing quality?
This is the story of how we rebuilt our entire video production setup to do exactly that.
Where It All Started
Back in 2018, when we got our first Tesla Model 3, we weren’t even a business yet.
We were just creating content.
There wasn’t much out there for EV owners, so we focused on one thing:
Provide value.
That mindset stuck. Even today, as a full business, our core goal hasn’t changed:
→ Educate
→ Simplify
→ Help people do it themselves
But as we grew, one big problem kept coming up…
The Bottleneck: Content Production
We wanted to produce more content.
Courses. Weekly YouTube videos. Tutorials. Social clips.
But traditional production setups slowed us down:
- Multiple cameras (3+ setups)
- Manual switching
- Heavy editing workflows
- Needing multiple people to film
Even with interns or help, it was hard to manage.
We kept asking:
What if one person could walk in, press a button, and just start filming?
The Shift: AI + PTZ Cameras
That’s where we started exploring PTZ cameras (pan-tilt-zoom).
These are remotely controlled cameras—but now, with AI tracking, they’ve become something way more powerful.
We ended up using AI-tracking cameras connected via NDI (Network Device Interface), which means:
- Everything runs through our network
- No traditional HDMI routing
- Near-zero latency
- Instant access from any computer
This became the foundation of our system.
Our Setup (Simplified)
Here’s how everything connects:
1. Network-Based Cameras (NDI)
All cameras are plugged into a high-speed network switch.
- Runs on UniFi network
- 1G–10G speeds
- Instant camera feeds into OBS
No capture cards. No complicated routing.
2. OBS Studio as the Control Hub
We use OBS Studio to:
- View all camera feeds
- Switch angles live
- Record everything simultaneously
Everything is centralized and controlled from one place.
3. OBSBOT Tail 2 – AI Tracking Cameras
These cameras:
- Follow movement automatically
- Zoom in for close-ups
- Maintain framing without a camera operator
This is huge for DIY content where details matter.
4. Custom Rail System (DIY Build)
Instead of expensive studio rigs, we built our own rail system:
- Mounted to warehouse beams
- Uses reinforced rolling mounts
- Supports multiple cameras
Not pretty—but extremely effective.
5. Wired Over Wireless (On Purpose)
We tested wireless setups…
Didn’t like them.
So we went fully wired:
- Cat6 Ethernet
- No signal drops
- No interference
- Consistent performance
Reliability > convenience.
Eliminating One of the Biggest Problems: Cables
One of the biggest issues in any studio?
Cables on the floor.
They’re:
- Dangerous
- Annoying
- Constantly in the way
So we:
- Routed everything overhead
- Used tension and zip systems
- Kept cables completely off the ground
Simple fix. Huge quality-of-life improvement.
The Game-Changer: Self-Editing While Filming
This is where things get really interesting.
We set up a foot-controlled switch system that connects via Bluetooth.
Using a device like the AirTurn Quad 500, we mapped:
- Camera 1
- Camera 2
- Camera 3
- Camera 4
So while filming, we can:
→ Switch camera angles
→ Control the recording
→ Edit in real time
No editor required during filming.
Remote Control Everything
Because everything is network-based:
- Cameras can be controlled remotely
- OBS can be triggered via webhooks
- Recording can start with a single command
We even built a system where:
→ One button triggers recording across all cameras
→ Everything syncs automatically
In theory…
Someone could run this entire setup remotely.
What’s Next: Fully Automated Editing
This is where AI really comes in.
We’re working toward:
- AI reads our past editing patterns (XML/project data)
- Auto-edits the video
- Outputs a near-finished version
- Team reviews and approves
The goal:
Record → Auto Edit → Approve → Publish
Why This Matters
As we expand into more products and content, we don’t just need more videos.
We need:
- Faster production
- Less friction
- Scalable systems
This setup forces us to rethink everything:
- How we film
- How we edit
- How we publish
And honestly…
We’re just getting started.
Final Thoughts
This is Version 1.
It’s not perfect. It’s a little scrappy. But it works.
And more importantly:
It unlocks scale.
We can now:
- Film faster
- Reduce team dependency
- Create more content consistently
And continue doing what we’ve always done since day one:
Provide value.
