---
title: "Research & Systems Engineering — Daniel Herrington"
description: "Systems research papers and warehouse-scale ML infrastructure analyses by Daniel Herrington."
url: "https://herrington.ai/research/"
markdown_url: "https://herrington.ai/research/index.md"
author: "Daniel Herrington"
date: "2026-10-31"
updated: "2026-10-31"
markdown_tokens: 408
content_signal: "ai-train=yes, search=yes, ai-input=yes"
mcp_endpoint: "https://herrington.ai/mcp"
---
# Research & Systems Engineering — Daniel Herrington

Distributed ML training efficiency, ML Productivity Goodput, megakernels, and accelerator hardware co-design.

## Publications

- **2026-09-12** — [Beyond Peak TFLOPs: Why Communication-Compute Overlap Dictates ML Productivity Goodput](https://herrington.ai/research/beyond-peak-tflops/) ([.md](https://herrington.ai/research/beyond-peak-tflops/index.md)): A systems-engineering analysis responding to Cursor's Mixture-of-Kittens (MoK) paper. Analyzes how unified megakernels, pull-based dispatch, and hardware co-design redefine MoE training goodput across NVIDIA Blackwell and Google TPU architectures.
- **2025-02-10** — [ML Productivity Goodput in LLM Training (arXiv:2502.06982)](https://arxiv.org/abs/2502.06982): Google research paper defining cluster efficiency metrics, downtime attribution, and training goodput for massive AI models.
- **2024-10-15** — [Measuring ML Productivity with Goodput (Google Cloud Blog)](https://cloud.google.com/blog/products/ai-platform/goodput-metric-as-measure-of-ml-productivity): Production architectural overview of tracking training productivity across TPU v4/v5e pods.