Gimlet Labs
Heterogeneous neocloud for AI workloads, decoupling inference from underlying hardware via compilers and orchestration.
About
Gimlet Labs is an applied research lab working on the systems layer underneath AI workloads. Two current projects: Gimlet cloud, serverless inference for AI agents, and kforge, which autonomously generates optimized low-level kernels directly from PyTorch. Research areas include kernel generation, datacenter scheduling, edge/cloud orchestration, AI compilers, and hardware architectures.
The stated goal is to make AI workloads 10x more efficient by expanding the pool of usable compute and improving how it's orchestrated.
Raised an $80M Series A in March 2026 led by Menlo Ventures, with Eclipse, Factory, Prosperity7, and Triatomic. Based at 255 Potrero Ave in San Francisco.