# D2

Published articles for D2.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## PCBRouteAgent: How Close Are LLM Agents to Becoming PCB Routing Engineers?

DevFeed: [PCBRouteAgent: How Close Are LLM Agents to Becoming PCB Routing Engineers?](<https://devfeed.tech/articles/pcbrouteagent-how-close-are-llm-agents-to-becoming-pcb-routing-engineers-120291.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.08175>)

Author: Yue Zhang, Jilei Sun, Liqiang Jing, Yijie Wei, Ruohan Yang

Published: 2026-11-06T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [kicad](<https://devfeed.tech/topics/kicad.md>), [harvestbench](<https://devfeed.tech/topics/harvestbench.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [best](<https://devfeed.tech/tags/best.md>), [board](<https://devfeed.tech/tags/board.md>), [bounded](<https://devfeed.tech/tags/bounded.md>), [can](<https://devfeed.tech/tags/can.md>), [circuit](<https://devfeed.tech/tags/circuit.md>), [combines](<https://devfeed.tech/tags/combines.md>), [complex](<https://devfeed.tech/tags/complex.md>), [connectivity](<https://devfeed.tech/tags/connectivity.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [d2](<https://devfeed.tech/tags/d2.md>), [d3](<https://devfeed.tech/tags/d3.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-agents](<https://devfeed.tech/tags/llm-agents.md>), [pcb](<https://devfeed.tech/tags/pcb.md>), [pcb-routing](<https://devfeed.tech/tags/pcb-routing.md>), [routing](<https://devfeed.tech/tags/routing.md>)

### AI overview

PCBRouteAgent pairs board-level planning with local routing execution: its Global Routing Planner uses board images and structured electrical and geometric data to choose connection topology, routing order, corridors, and layer transitions. A Detailed Routing Executor implements those guides through KiCad's native routing engine, attempts bounded local recovery, and returns unresolved failures as structured feedback for targeted guide revisions while preserving accepted routes. Across 247 synthetic and real-world boards, reported clean-pass rates were 100% on D2, 94.9% on D3-A, 100% on D3-B, and 40% on D3-C. On the three real-world subsets, it exceeded the best evaluated baseline by 13.9, 36, and 10 percentage points, respectively. Results vary substantially across test splits.

### Source excerpt

arXiv:2611.08175v1 Announce Type: new Abstract: Automating printed circuit board (PCB) routing requires coordinating board-level connectivity decisions with geometric execution under strict design rules. Although recent agents can interact with native routing tools, reliable routing on complex boards remains challenging. We present PCBRouteAgent, a multimodal PCB routing agent inspired by how human engineers coordinate global strategy with local execution. A Global Routing Planner combines board images and structured electrical and geometric information to specify connection topology, routing order, corridors, and layer transitions. A Detailed Routing Executor realizes these guides through KiCad's native routing engine and performs bounded local recovery. Unresolved failures are returned as structured feedback, enabling targeted guide revisions while preserving accepted routes. We evaluate PCBRouteAgent on 247 synthetic and real-world boards from four PCBWorld-Bench test splits. It achieves clean-pass rates of 100%, 94.9%, 100%, and 40% on D2, D3-A, D3-B, and D3-C, respectively. On the three real-world subsets, PCBRouteAgent outperforms the best evaluated baseline in clean-pass rate by 13.9, 36, and 10 percentage points, respectively.