£350k · Pre-Seed · AI · Bristol, UK
Bristol-based Nearhuman has raised £350,000 in pre-seed funding led by SFC Capital to develop on-device safety technology for shared e-scooter and e-bike fleets. The company says the capital will support product development, on-street testing and commercial deployment, while its first live pilot is running in Bristol.
The Nearhuman team at Bristol Robotics Laboratory. Image: Nearhuman.
Retrofit hardware lowers the first deployment hurdle
Nearhuman's camera module is designed to fit vehicles already in service. Computer vision running on the vehicle detects pavement riding, pedestrian proximity, near-misses and road-surface defects. The company says the system operates at 30 frames per second on roughly 2 W of power and processes and redacts footage locally, limiting the need to send sensitive video to the cloud.
That architecture addresses two practical adoption barriers. Fleet operators do not need to replace working scooters or bikes to add the system, and local processing reduces dependence on continuous connectivity. The retrofit model can make a pilot cheaper and faster to start, but it also puts a high standard on installation, power use and reliability across weather, vibration and changing street conditions.
The Bristol pilot is designed to test those conditions. Approved operators, councils, Department for Transport users and research partners can access a dashboard showing geotagged events, photographs and confidence scores. This is a live test rather than evidence of a citywide deployment or paid commercial contract, so the relevant milestone is whether the data changes operational decisions.
Pilot data has to become an operator workflow
For a micromobility operator, detecting a risky moment is only the first step. The product becomes more valuable if recurring events can be turned into rider interventions, fleet changes or evidence for discussions with city authorities. For councils, the same vehicles could create a street-level view of road, cycle-lane and pavement conditions without a separate survey fleet.
That gives Nearhuman a route from a hardware module to an ongoing data product, but only if customers trust the detections and can integrate them into existing safety work. False alerts create review costs; missed events weaken the case for using the data. On-device redaction can help with privacy, while the pilot's controlled access gives Nearhuman a setting to learn which outputs different stakeholders will actually use.
Founder Faizan Mir developed the company after studying robotics at UWE Bristol and working from Bristol Robotics Laboratory. SFC Capital principal Ed Stevenson describes e-scooters as Nearhuman's starting market for a broader physical-AI platform. The company says it plans further products, but the near-term commercial proof remains narrower: demonstrate that the Bristol pilot produces reliable, actionable safety data and converts into repeatable operator deployments.
