CEO, Ambient Scientific Inc.
Gajendra Prasad (GP) Singh is CEO of Ambient Scientific Inc. which is a Deep Learning
Semiconductor startup working on Extremely Low Power AI-inference chip products. Under his
leadership, the company has developed breakthrough Digital Analog (DigAn™) AI computing
technology that realizes advantages of both Analog and Digital, simultaneously. DigAn™ brings
to life the promises of Analog AI computing i.e. high performance, low power and low cost
while still retaining the advantages of digital chips i.e. accurate, manufacturable, flexible and
scalable. Ambient Scientific company has unveiled first product in its roadmap – an Ultra-Low
Power AI processor for On-Device AI computing for a wide variety of IoT applications such as
Natural Language Processing, anomaly detection, industrial applications, smart sensor
applications etc. The company plans to add AI-inference solutions for vision and Edge Cloud
solutions in future.
Highlights of his career:
An engineer at heart. An entrepreneur, Innovator corporate and technical manager by
profession. Always leading ground breaking semiconductor innovations. 50+ chip tape outs
including deep sub-micron technology nodes (all the way to 7nm). Able to contribute at many
levels in org-chart from executive role to a hands-on-engineer. 50+ patents. 5+ publications.
He started his career 25+ years ago as a design engineer and slowly learnt skills to lead
innovations and teams in various big companies and start-ups before founding Ambient
Digital Analog (DigAn™) AI Computing: Promises of Analog AI with strength of Digital CMOS
Ambient Scientific has invented industry’s first Digital Analog (DigAn™) AI technology. DigAn™ AI is an advanced version of On-Chip Analog AI computing technology that provides ultra-high performance and efficiency of Analog AI with implementation and commercial advantages of Digital chips. The company demonstrated its innovative DigAn technology with its Industry’s lowest power On-Device AI solution for ‘Wake on Word’ and other On-Device AI applications.
Ambient’s DigAn™ AI Processor computes an entire layer of neurons in a single clock cycle using custom Analog techniques, a task that takes several hundreds or even thousands of clock cycles in traditional implementations. DigAn™ uses standard digital transistors to implement its analog AI compute functions.Analog is a great fit for AI computing. However, prior implementations had severe limitations in accuracy and computing range – two important parameters for optimal AI solution. Ambient has eliminated these limitations. While prior implementations have struggled to achieve 8-bit accuracy, Ambient’s DigAn™ (Digital Analog) technology can achieve 32-bit accuracy. This is 16 million times more accurate in computer language.
Ambient’s first solution is targeted to Industry’s lowest power On-Device ‘Always-ON’ AI for ‘Wake on Word’ and Anomaly Detection applications. It uses only 80μW (microwatts) power or less for AI computing, which is 200x lower than industry standard and 2x-3x lower than recent advancements. In addition, Ambient has invented important components such as Ultra-Low Power Analog to Digital Converter (ADC) to ingest sensor data and Ultra Low Power clock generator (LPO) to enable chip operation without external host.
Founded in 2017, Ambient Scientific’s mission is to enable a portfolio of AI solutions that scales from tiny devices to large computer systems. The company’s grass root level innovations deliver high performance, ultra-low power DigAn™ neural processors, that are easy to program. The company’s ambition is to be #1 AI computing solution provider of the world. Ambient Scientific is headquartered in San Jose, California.