Design Conductor 2.0: An agent builds a TurboQuant inference accelerator in 80 hours
Research area: Computer architecture Authors: The Verkor Team, Ravi Krishna, Suresh Krishna, David Chin Published: 2026-05-06 arXiv: 2605.05170
Driven by a rapid co-evolution of both harness and underlying models, LLM agents are improving at a dizzying pace. In the authors' prior work (performed in Dec. 2025), they introduced Design Conductor (or just Conductor), a system capable of building a 5-stage Linux-capable RISC-V CPU in 12 hours. In this work, they introduce an updated multi-agent harness powered by frontier models released in April 2026, which is able to handle 80x larger tasks, at higher quality, fully autonomously.
Following a brief introduction, the paper examines 4 designs that the system produced autonomously, including VerTQ, an LLM inference accelerator which hard-wires support for TurboQuant in a 240-cycle pipeline, starting from the TurboQuant arXiv paper. VerTQ includes heavy compute processing, with 5129 FP16/32 units; the design was mapped to an FPGA at 125 MHz and consumes 5.7 mm² in TSMC 16FF (8 attention pipes).
The authors review the key new characteristics that enabled these results. Finally, they analyze Design Conductor's token usage and other empirical characteristics, including its limitations.
--- *Auto-collected on 2026-05-08*