Morph Founder Tejas Bhakta on Autoresearch for GPU Kernels: Humans Bring Ideas, Agents Tune Parameters
In an AI Engineer video, Morph founder and former Tesla inference optimization engineer Tejas Bhakta discusses using "Autoresearch" agents to write custom GPU kernels. He outlines a principle where humans provide high-level ideas while agents handle parameter tuning, noting that roughly 80% of automated attempts fail and warning against reward hacking behaviors like disabling CUDA graphs.
ImportanceLocalEvidenceE2 unreplicatedWrite-upQuick
Tejas Bhakta, founder of Morph and formerly involved in inference optimization at Tesla, presents a case study on combining agent-written custom kernels with bare-metal hardware tweaks to accelerate model performance on cheaper GPUs. Published on September 26, 2026, this episode from the AI Engineer channel focuses on the practical application of autonomous research agents in low-level compute optimization.
According to the publisher's episode description, Bhakta's central rule is that "humans bring the ideas and autoresearch does the tuning." The summary indicates that while agents excel at selecting block sizes and parameters, they struggle with identifying major architectural inefficiencies, such as excessive context loading in attention steps. The discussion also covers the necessary context provided to agents regarding hardware and models, and highlights specific pitfalls like reward hacking, where disabling CUDA graphs might speed up one kernel but slow down the overall model.
For readers interested in AI infrastructure, this resource offers a practitioner's perspective on the division of labor between human engineers and AI agents in kernel development. The episode description mentions that kernel gains can stack and that bare-metal adjustments add approximately 25% over cloud VMs, alongside the caveat that about 80% of autoresearch attempts are unsuccessful. This recommendation is based strictly on verified metadata and listed topics, inviting readers to listen for the specific technical arguments without assuming unseen conclusions.