- optoML has successfully completed its 12nm tapeout with TSMC, marking a critical execution milestone in advancing its proprietary semiconductor architecture.
- The company’s patented analog in-memory compute (AiMC) architecture delivers up to 50× higher energy efficiency compared to conventional digital accelerators, making it ideally suited for AI workloads across edge devices, enterprise systems, and hyperscale data centers.
- To accelerate commercialization, optoML has signed a Memorandum of Understanding (MoU) with Kaynes Semiconductor to support wafer assembly, testing, and scalable packaging once production begins.
- The newly raised funds will be deployed to expand hiring, strengthen engineering capabilities, and initiate development of next-generation chips, building on the successful TSMC tapeout milestone.
Bengaluru/Chennai | February 24, 2026: optoML, a fabless semiconductor startup pioneering Analog-in-Memory Compute (AiMC) with optical interconnects, has raised $1.8 million in a Pre-Series A round led by Bluehill.VC and A99. The transaction is pending regulatory approval.
The capital will be used to scale hiring, initiate work on next-generation chips, and advance commercialization following the successful 12nm tapeout with TSMC.
Technology Breakthroughs
- optoML has completed 12nm TSMC tapeout, realizing its core AiMC architecture and SoC integration path.
- The company’s patented in-memory compute architecture delivers up to 50× higher energy efficiency compared to traditional digital accelerators.
- Designed for edge, enterprise, and data center AI workloads, optoML’s platform reduces latency, silicon area, and cost.
- optoML has signed an MoU with Kaynes Semiconductor to support assembly and testing once wafers arrive from TSMC, accelerating the journey from silicon to scalable packaging, manufacturing, and product readiness.
Manu Iyer, General Partner, Bluehill.VC, said: “India’s deep-tech ecosystem is coming into its own, and semiconductors are at the forefront. optoML sits at the intersection of two irreversible shifts in compute, analog in-memory architectures and optical interconnects. Their 12nm tapeout with TSMC and partnership with Kaynes mark a critical transition from research to real silicon and scalable production. This is the kind of deep-tech platform that can reshape the cost and efficiency curve of AI infrastructure.”
Vignesh S, General Partner, A99, added: “As AI workloads move to the edge, power efficiency becomes the real constraint. optoML’s analog-in-memory approach addresses exactly that, shifting how AI inference can be delivered at lower energy and cost. Foundational semiconductor IP in AI infrastructure will matter immensely over the next decade, and this investment aligns with our thesis at A99.”
Saravana Maruthamuthu, Founder & CEO, optoML, said: “We are excited to have Bluehill and A99 partner with us. Their conviction and deep experience in early-stage deep tech investments will guide our path from test chip success to commercial viability, customer deployments, and scaled production. This funding puts us on track for productization, qualification, and real-world impact, turning efficient AI compute into deployable systems.”
About the Founder
Saravana Maruthamuthu brings 17 years of semiconductor and systems experience, including a decade in Germany. His career spans Intel, Qualcomm, Continental, and Isar Aerospace, across advanced nodes, analog mixed signal, and production-grade platforms.
About Bluehill.VC
- Chennai-based Bluehill.VC manages a ₹350 crore deep-tech fund focused on IP-led engineering and science domains.
- Investment areas include Energy, Space, Defence, Semiconductors, Material Science, Water, Manufacturing, Robotics, Industrial IoT.
- Average investment size ranges from $750K to $2M, with up to $3M in follow-on rounds.
About optoML
optoML is a pioneering semiconductor company delivering the next evolution in AI compute. Headquartered in Singapore with an R&D center in Bengaluru, optoML is developing the world’s first FinFET-based analog AI SoC platform with integrated optical interconnects.
Key offerings include:
- Scalable NPU (Neural Processing Unit).
- Full software stack.
- Chip-to-chip optical links.
Its patented architecture delivers up to 50× higher energy efficiency, making it ideal for edge, enterprise, and data center AI workloads.
