India’s semiconductor ecosystem continues to gather momentum as deep-tech startup optoML secures $1.8 million in a pre-Series A funding round led by Bluehill.VC and A99. The fresh capital marks a critical milestone for the fabless chip design company, which focuses on building energy-efficient AI System-on-Chip (SoC) platforms powered by analog-in-memory compute architectures and optical interconnects.
The funding follows the successful completion of optoML’s 12 nm tapeout with TSMC, the world’s leading semiconductor foundry. With silicon now in fabrication, the startup moves from design validation into execution and commercialization.
Capital Fuels Hiring and Next-Gen Development
optoML plans to channel the newly raised capital into two immediate priorities: expanding its engineering team and accelerating development of its next-generation AI chips. The company aims to deepen expertise across chip architecture, analog design, photonics integration, firmware, and AI model optimization.
Founder Saravana Maruthamuthu leads optoML with a vision to rethink AI hardware from the ground up. Instead of relying on conventional digital accelerators that separate memory and compute units, optoML integrates compute directly within memory arrays. This approach dramatically reduces data movement, which remains one of the largest sources of power consumption and latency in AI workloads.
By strengthening its team and moving quickly on the next iteration of its architecture, optoML seeks to position itself ahead of global competition in energy-efficient AI hardware.
Analog-in-Memory Compute: A Shift in AI Architecture
Traditional AI accelerators rely on digital logic and shuttle data repeatedly between memory and compute units. That movement consumes enormous energy, especially in data centre and edge inference workloads.
optoML eliminates this bottleneck by embedding computation within memory itself. The company’s patented analog-in-memory design performs matrix operations directly where data resides. This architecture cuts unnecessary transfers and slashes energy consumption.
The company claims its technology delivers up to 50x higher energy efficiency compared to conventional digital accelerators. That improvement addresses one of the biggest challenges in modern AI: power constraints.
As AI adoption expands across industries, energy efficiency determines scalability. Hyperscale data centres struggle with rising electricity costs. Edge devices face strict thermal and battery limitations. Enterprises demand cost-effective inference systems. optoML targets all three segments with a unified, scalable SoC platform.
Optical Interconnects Enhance Performance
Beyond analog-in-memory compute, optoML integrates optical interconnects into its architecture. Optical communication offers significant advantages over traditional electrical interconnects, especially in bandwidth density and signal integrity.
Electrical interconnects face resistance, heat generation, and signal degradation as workloads scale. Optical interconnects transmit data using light, enabling higher speeds and lower latency across chiplets or system modules.
By combining analog-in-memory processing with optical data transfer, optoML creates a hybrid architecture optimized for modern AI pipelines. This integration strengthens performance in large model inference, real-time analytics, and high-throughput enterprise workloads.
12 nm Tapeout with TSMC Marks Technical Milestone
The company recently completed its 12 nm FinFET tapeout with TSMC. Tapeout marks the stage where engineers finalize the chip design and send it for fabrication. This milestone signals design maturity and readiness for silicon validation.
FinFET nodes offer improved control over current leakage and power efficiency compared to planar transistor technologies. By building on a 12 nm FinFET process, optoML balances performance, cost, and scalability.
Once wafers return from fabrication, optoML will begin testing, characterization, and performance benchmarking. These steps will validate energy efficiency claims and establish reliability across operating conditions.
Partnership with Kaynes Semiconductor
To streamline post-fabrication processes, optoML signed a memorandum of understanding with Kaynes Semiconductor. The partnership will support assembly and testing once wafers arrive from TSMC.
Assembly, packaging, and testing play a critical role in semiconductor commercialization. Advanced packaging ensures signal integrity and thermal stability. Rigorous testing validates yield, durability, and consistency.
By collaborating with Kaynes Semiconductor, optoML strengthens its supply chain strategy within India. This alignment supports broader national ambitions to build a resilient semiconductor ecosystem that spans design, manufacturing, and testing.
Fabless Strategy Enables Scalability
optoML operates as a fabless semiconductor company. It focuses on architecture, design, and system innovation while leveraging leading foundries for fabrication.
This model enables flexibility and capital efficiency. The company avoids the enormous infrastructure costs associated with building fabrication plants. Instead, it allocates resources toward R&D, talent acquisition, and ecosystem partnerships.
Fabless companies drive much of today’s semiconductor innovation. optoML’s approach allows rapid iteration and faster alignment with evolving AI workload demands.
Market Opportunity Across Edge and Data Centres
AI hardware demand continues to surge across three major segments: edge devices, enterprise infrastructure, and hyperscale data centres.
Edge computing requires compact, low-power AI accelerators that handle inference locally. Devices such as industrial sensors, smart cameras, and autonomous systems need real-time processing with strict energy limits. optoML’s energy-efficient SoCs offer a compelling solution for such environments.
Enterprise deployments seek AI acceleration for analytics, cybersecurity, and automation. Cost efficiency and scalability drive purchasing decisions in this segment. Analog-in-memory compute reduces power bills and cooling requirements, strengthening total cost of ownership metrics.
Data centres represent the largest opportunity. AI model sizes continue to grow, and inference workloads run around the clock. Energy efficiency directly impacts operational expenses. optoML’s claimed 50x efficiency improvement could translate into significant cost savings at scale.
Bluehill.VC Backs Deep-Tech Vision
Lead investor Bluehill.VC focuses on deep-tech sectors including semiconductors, defence, energy, and industrial technology. The firm supports startups that tackle complex engineering challenges with defensible intellectual property.
By backing optoML, Bluehill.VC signals confidence in analog-in-memory compute and optical integration as future pillars of AI hardware. Deep-tech investors typically prioritize long development cycles and high entry barriers. optoML’s patented architecture and advanced node tapeout align with that thesis.
The participation of A99 further strengthens the investor base and broadens strategic guidance for scaling operations.
Strengthening India’s Semiconductor Ambitions
India’s semiconductor sector gains strategic importance as global supply chains diversify. The government pushes initiatives to support chip design and manufacturing capabilities. Startups like optoML contribute critical intellectual property and architectural innovation to this ecosystem.
By combining global fabrication partnerships with domestic assembly collaborations, optoML creates a hybrid model that integrates international excellence with local capability development.
The company’s focus on AI SoCs also aligns with global demand trends. Nations seek technological leadership in artificial intelligence, and efficient hardware forms the backbone of that ambition.
Road Ahead
The next twelve months will prove decisive for optoML. Silicon validation will determine performance benchmarks and customer traction. The team expansion will influence execution speed. Strategic partnerships will shape go-to-market pathways.
If silicon testing confirms the promised efficiency gains, optoML could attract further institutional investment and strategic alliances. Hyperscalers, enterprise integrators, and edge device manufacturers may explore pilot deployments.
Saravana Maruthamuthu and his team now stand at a pivotal stage. They have secured capital, achieved tapeout, and established supply chain partnerships. The focus shifts toward delivering measurable results and building commercial momentum.
In an AI-driven world where power efficiency defines scalability, optoML aims to reshape how chips process intelligence. With $1.8 million in fresh funding and a validated 12 nm design in motion, the startup enters its next chapter with ambition and technical depth.
Also Read – Why AI Wrappers Won’t Survive Long-Term