Beyond the Demo Video
Robotics is a field where a polished demo can hide a fragile system. Our own test rigs and autonomy benchmarks — built specifically to validate what robotics companies claim — shape a framework we apply to every opportunity we evaluate.
Hardware Maturity
- Is the mechanical design suited to the target environment, or optimized for a controlled demo?
- How does the system degrade — gracefully, or catastrophically — outside its training distribution?
- What's the real cost and complexity of manufacturing at scale?
The Autonomy Stack
- Perception, planning, and control each fail differently. We look at where a team's stack is genuinely differentiated versus assembled from off-the-shelf components.
- Simulation results are a starting point, not a conclusion. We weight real-world deployment data heavily.
Path to Deployment
- Who is the first paying customer, and what does "good enough" look like for them?
- Is the unit economics story realistic once you account for maintenance, downtime, and integration?
Why We Build Our Own Test Rigs
Our own manipulation and perception benchmarks — built to stress-test other companies' claims, not to produce a product of our own — give us a reference point for these questions that a purely financial read of a pitch deck can't provide.
What Happens Next
When a company clears this framework, we don't just consider investing ourselves — we bring the opportunity to our network of outside investors who can invest directly.
Quant Labs View
What matters technically: how a system degrades outside its training distribution — gracefully or catastrophically — since that's usually the real gap between a demo and a deployed product.
What we would test: the manipulation or perception stack under conditions the company didn't choose for the demo, not a replay of their own test suite.
What could invalidate the thesis: a robotics stack assembled from off-the-shelf components dressed up as proprietary technology — common enough that it's a specific thing we check for.
Commercial implication: hardware maturity and deployment economics matter as much as the autonomy stack itself — a technically impressive robot that can't be manufactured or maintained at a viable cost isn't a fundable technology story.