Lola Vision Systems, an AI infrastructure company, has launched software designed to streamline the process of running AI models on specific hardware. The company's 'compiler toolchain' aims to reduce the manual setup time for AI models, which can currently take hundreds of hours. Lola Vision is also developing its own semiconductor chips to further automate this process.
The development of more efficient AI model deployment tools is critical for the widespread adoption of AI on edge devices, potentially lowering costs and increasing performance for mission-critical applications.
Lola Vision Systems, an AI infrastructure startup, has launched a software solution aimed at simplifying the deployment of AI models on specialized hardware. Founded by Tayo Adesanya, who previously worked with microchips and AI processors for large manufacturers, the company seeks to address a significant bottleneck in the AI computing market.
Adesanya stated that manually setting up an AI model on new hardware can take around 200 hours for initial testing. Lola Vision's core product is a "compiler toolchain" software that translates AI models and client code into instructions executable by a specific chip. The company is also developing its own semiconductor chips to further automate this process.
According to Lola Vision, its solution offers faster setup times, lower power consumption, and improved accuracy and reliability for mission-critical applications in sectors like aerospace. The company claims its software can help clients run more accurate models on their own data, which is crucial for regulatory review and field reliability.
Lola Vision positions itself as an alternative to existing solutions like NVIDIA's Jetson modules, which Adesanya claims often require extensive setup and debugging, leading to performance issues and exceeding edge computing power budgets. The company has secured one customer and has letters of intent from a dozen more for its upcoming chips. To generate revenue sooner, Lola Vision will also license its software for use on existing hardware. To date, the company has raised just over $1 million in funding.
Pick the topics you care about. Get only what matters, on your cadence.
