Phyll
Setting the Scene
Potholes, rusting hydrants, and cracked sidewalks often go unnoticed until they cause costly disruptions, dangerous accidents, or insurance claims.
Meanwhile, billions of hours of video are captured every week by dash
cams, CCTV, and drones, yet this data remains untapped as a source of real-time infrastructure intelligence.
This week’s company turns everyday footage into a live, searchable 3D map of infrastructure, giving cities, communities, and organizations the visibility they need to monitor conditions and prevent failures.
In a Sentence
Phyll is a real-time AI platform that transforms footage from dash cams, CCTVs, and drones into a searchable 3D map, giving cities, organizations, and insurers visibility into their infrastructure.
Real-time AI: Detects, classifies, and tracks infrastructure assets while assessing condition and risk in real time.
Footage: Leverages pre-existing video streams, turning passive footage into active intelligence.
Visibility: Organizes data into a natural language searchable view of infrastructure health to prevent failures and plan proactively.
Bulleted Version:
- Similar to how Google Maps tracks traffic in real time, Phyll tracks the status of infrastructure using footage from dash cams, CCTVs, and drones.
The Basics
- Headquarters: Philadelphia, PA
- Employee Count: 3
- Business model: B2B/B2G SaaS; recurring per-asset and per-mile licensing, with video processed at <$0.001 per asset
- Early traction: Paid pilots in Austin and London, active micropilot in Philadelphia, $1.25M qualified pipeline
Due Diligence
WHAT WE LIKE
Market Opportunity: Governments, insurers, and utilities spend billions fixing infrastructure failures, but little on preventing them; Phyll shifts the spend toward proactive monitoring and smarter planning.
Data Monetization: By enabling drivers and communities to monetize existing dash cam, CCTV, and drone footage, Phyll taps into pre-existing streams rather than relying on expensive new hardware deployments.
Network Flywheel: Each mile of captured footage makes the system smarter: drivers, drones, and cameras feed into a self-reinforcing loop of accuracy, coverage, and adoption.
POTENTIAL RISKS
Operational Adoption: City and utility procurement cycles are notoriously slow, which could delay scaling even with strong technical performance.
Data Variability: Footage quality from dash
cams, CCTVs, and drones may vary widely, creating challenges in standardizing outputs and ensuring accuracy.
Coverage Gaps: Less popular areas or lightly traveled routes may generate limited footage, making it harder to deliver consistent monitoring across entire regions.
Founder Profile
Ethan Berg, CEO: Background in 3D spatial computing, previously founded a company to make immersive tech accessible.
Comps
Cyvl: Hardware that attaches to cars to focus predominantly on road repair.
Blyncsy: Produces 2D snapshots of road-level detections (potholes, paint, signage) using dash cams and Google Street View; mostly used for DOT compliance.
Blackshark.ai: Satellite-first mapping at global scale, but not optimized for detailed ground-level infrastructure monitoring.
Why Phyll
By tapping into existing video, scaling through community contribution, and compounding insights with every pass, this company fills the cracks in infrastructure monitoring.
