Runtime Engineer

Lemurian Labs raises $28M to expand its software-first approach for AI at scale. SANTA CLARA, Calif., Dec. 3, 2025 - Lemurian Labs today announced it has secured $28 million in an oversubscribed Series A round, including capital previously raised from convertible securities. Lemurian is developing a software-centric, hardware-agnostic solution that enables AI workloads to run efficiently on any hardware, at any scale. By treating the entire system as a unified compute fabric, Lemurian simplifies AI development, allowing developers to write code once and deploy it seamlessly across edge, cloud and on-premise environments. This new approach delivers faster deployment, greater flexibility and lower infrastructure costs at scale. "Scaling AI is the next frontier, but that's not possible on platforms designed for yesterday's workloads," said Jay Dawani, co-founder and CEO of Lemurian Labs. "For decades, faster chips delivered 'free gains,' but now the real bottleneck is software. Lemurian is rebuilding the software stack from the ground up to eliminate vendor lock-in, control costs and give developers the flexibility to run AI anywhere on their terms." Pebblebed Ventures and Hexagon co-led the round, with participation from Oval Park Capital, which led Lemurian's 2022 seed round, as well as Origin Ventures, Blackhorn Ventures, Uncorrelated Ventures, Untapped Ventures, Planetary Ventures, 1Flourish Ventures, Animal Capital, Stepchange VC and Silicon Catalyst Ventures. Lemurian Labs' founders and leadership team hail from industry leaders such as NVIDIA, Qualcomm, Sun Microsystems, IBM and Intel. "Lemurian is reframing the grim choice that AI's hardware-software interface has forced on users: choosing between vendor-locked vertical stacks or brittle, rewrite-prone portability," said Keith Adams, founding partner at Pebblebed Ventures. "Jay and his team bring technical virtuosity that lets you run your AI code as written, on whatever hardware makes sense - full stop, no compromises." Pebblebed is an early-stage venture firm by and for builders. It works closely with a small group of founders who are turning technical secrets into generational companies. Vendor-locked software doesn't just stifle innovation; it drives up costs and wastes energy. AI workloads are projected to consume 20% of global electricity by 2030-2035, a trajectory accelerated by inefficient, proprietary software. The current model, built on closed, vertically integrated stacks, simply cannot scale fast enough to meet the pace of AI advancement. Lemurian Labs' open, software-first approach optimizes performance across heterogeneous hardware, from GPUs to edge devices, helping organizations run AI more efficiently and responsibly at scale. "Everyone in AI wants to see healthy competition in the GPU market to accelerate innovation," said Salil Deshpande, general partner at Uncorrelated Ventures. "But in order for that to happen, someone has to develop CUDA-like software for a wide range of GPUs and other processors, which is difficult; it's why I was excited to invest in Lemurian Labs." Lemurian will use the new funding to expand its engineering team, accelerate product development and deepen collaborations with ecosystem partners focused on sustainable compute and open AI innovation. For more information about Lemurian Labs and its technology, please visit lemurianlabs.com. About Lemurian Labs Lemurian Labs has created a universal platform that works across any hardware and spans compiler technology and runtime orchestration. This streamlined, hardware-agnostic approach simplifies AI development, enabling organizations to write code once and deploy it seamlessly across edge, cloud and on-premise environments. By cutting through the complexity of fragmented software ecosystems, Lemurian Labs empowers businesses to build and scale AI efficiently, affordably and responsibly - unlocking the full potential of AI innovation without proprietary constraints. Dec. 3, 2025 - Efficiency has emerged as a critical challenge for organizations deploying AI. While... SUNNYVALE, Calif. and TOKYO, Dec. 3, 2025 - NTT Research, Inc. and NTT R&D, divisions of... The new Mistral 3 family, spanning frontier-level to compact models, is optimized for NVIDIA platforms,... LAS VEGAS, Dec. 2, 2025 - AWS today announced a comprehensive expansion of its Nova... SAN SEBASTIÁN, Spain, Dec. 2, 2025 - As large language models continue to get larger and... Dec. 2, 2025 - As governments and large organizations seek to scale AI projects, some are...

Lemurian Labs has raised $28 million in a Series A round co-led by Pebblebed Ventures and Hexagon, with participation from seed investor Oval Park Capital and over half a dozen other backers. The artificial intelligence tooling startup develops an efficiency-optimised AI model compiler. AI models consist of kernels—code snippets designed to run in parallel across numerous graphics processing unit cores, enabling faster calculations than sequential processing. The funding comes three years after Lemurian's seed round. The company focuses on optimising how AI models execute on hardware infrastructure.

Lemurian Labs raises $28M Series A to scale its software-first solution that runs AI faster, more efficiently and on any hardware without vendor lock-in....