Bluehill.VC and A99 back optoML’s AI semiconductor vision
SoC semiconductor startup optoML has raised $1.8 million (approximately Rs 15 crore) in a pre-Series A funding round led by Bluehill.VC and A99. The fresh capital will be used to expand hiring and accelerate development of its next-generation chip platforms. The investment comes after the company successfully completed the tapeout of its 12 nm chip in collaboration with Taiwan Semiconductor Manufacturing Company (TSMC), marking a key milestone in its technology roadmap.
Building analog-in-memory architectures for AI workloads
Founded by Saravana Maruthamuthu, optoML develops analog-in-memory compute architectures integrated with optical interconnects to support artificial intelligence workloads. The company’s patented design enables computation directly within memory, reducing energy consumption and improving performance efficiency. According to optoML, its architecture can deliver up to 50 times higher energy efficiency compared to conventional digital accelerators, targeting applications across edge devices, enterprise infrastructure, and data centres.
Partnership with Kaynes to enable manufacturing and scale
optoML has signed a memorandum of understanding with Kaynes Semiconductor to support chip assembly and testing once wafers arrive from TSMC. The startup operates as a fabless semiconductor company, focusing on scalable AI System-on-Chip (SoC) platforms built on advanced FinFET process nodes. This partnership is expected to help streamline the transition from chip design to production and commercial deployment.
Positioning for growth in the AI semiconductor ecosystem
With rising demand for efficient AI hardware, optoML aims to establish itself as a provider of high-performance, energy-efficient compute platforms. Investors such as Bluehill.VC are actively backing deep-tech startups in semiconductors, defence, and industrial technologies, reflecting growing interest in India’s semiconductor innovation ecosystem.
Keep building. Keep learning. Keep growing with StartupByDoc.