CEOInterviews.AI
Start App
Try: AI strategy tariffs earnings guidance layoffs interest rates growth

Natalie Serrino

Co-founder, Gimlet Labs

Search every verified Natalie Serrino interview, podcast appearance, and on-the-record quote — each transcript cross-checked by AI and human review to confirm speaker identity. Natalie Serrino, co-founder of Gimlet Labs, discussed the company's work on a cross-vendor inference cloud in a May 2026 interview. She described Gimlet as building an inference cloud for agents that operates across heterogeneous hardware, tracing PyTorch workloads, segmenting them, and scheduling each part onto the most suitable chip from different vendors connected on a single fabric. Serrino stated that agentic inference is not a uniform workload, with different parts having different compute needs, and argued that connecting chips from different vendors in this way is a novel approach. She also noted that the company sells into frontier labs, sovereign clouds, and AI natives, and that its offering provides a mix-and-match environment for customers seeking to benefit from emerging architectures without building their own cloud infrastructure. Serrino also addressed hiring, saying Gimlet is recruiting across the tech stack, from low-level kernel engineering to higher-level software engineering, as the company builds an end-to-end cloud stack rather than a bare-metal service. She characterized the opportunity as large and the company as a "crazy growing ship," while a former Intel executive who joined Gimlet five months prior described being "blown away by the scale of the opportunity." Serrino emphasized that Gimlet's focus is on the software layer for orchestrating across hardware, a problem she said is set to compound.

Recent appearances

  • Gimlet's Cross-Vendor Inference Cloud

    May 12, 2026 · 📺 Semi Doped

    Gimlet Labs runs an inference cloud built on heterogeneous silicon. Their software traces a PyTorch workload, segments it into its component parts, and schedules each piece onto the best-suited hardware — connecting chips from different vendors on a single high-speed fabric. In this interview, Gimlet co-founder Natalie Serrino and former Intel executive Beltir walk through the architecture (graph trace, optimal split points, lowering each segment to TensorRT on NVIDIA and equivalents elsewhere), the three customer segments they sell into (frontier labs, sovereign clouds, AI natives), and a co…

    Watch on YouTube ↗

  • Optimizing AI Inference for Heterogeneous Clusters by Natalie Serrino, Founder @ Gimlet Labs

    April 22, 2026 · 📺 AI Performance Engineering

    Talk #0: Introductions and Meetup Updates by Chris Fregly and Antje Barth Talk #1: Optimizing AI Inference for Heterogeneous Clusters by Natalie Serrino, Founder @ Gimlet Labs This talk will cover the performance benefits and technical challenges of deploying inference workloads across heterogeneous hardware. It's a good fit for agents because agents are inherently heterogeneous, and combining GPUs with SRAM-centric architectures leads to major speedups for the same power envelope. But you also have to figure out how to slice workloads and orchestrate across all of this hardware, make the har…

    Watch on YouTube ↗

  • AI Kernel Generation: What's working, what's not, what's next – Natalie Serrino, Gimlet Labs

    December 17, 2025 · 📺 AI Engineer

    In this talk, we'll talk about how AI generated kernels can meaningfully speed up custom PyTorch code, without any human effort. Lots of great frameworks exist to optimize PyTorch with programmatic optimizations, such as Triton and MLX. But the strongest AI performance gains come from hand-written, low-level kernels that are targeted to the exact device and workload. These are tedious and time-consuming to write, especially when supporting multiple platforms. What if we could automate this process with AI? We'll cover the best practices for AI generating low-level kernels, from how to test a…

    Watch on YouTube ↗

  • Collaborative with Spencer Krause - E138 - Natalie Serrino (Software Engineer)

    February 26, 2024 · 📺 Collaborative With Spencer Krause

    Join me as I speak with Gimlet Labs co-founder Natalie Serrino about AI on the edge, computer vision, getting into computers and engineering, and more. If you enjoyed this episode and want to see / hear more like it please subscribe to Collaborative With Spencer Krause today. You'll get notified every time a new episode releases and it's the best way to support the channel! Gimlet is a developer tool that allows engineers to easily deploy AI pipelines on the edge. If you want to know more, please find Natalie on LinkedIn and mention CWSK when you reach out. Companies looking to outsource…

    Watch on YouTube ↗

The summary and quote tags on this profile are produced with AI assistance from verified, first-person interview transcripts, then checked by our team to confirm the speaker's identity and the accuracy of every quote. See how we verify →