Awesome AI Engineering Agents

English | 简体中文

English

An open, continuously updated map of AI agents, foundation models, papers, and open-source engineering tools for physical-product development. It follows AI adoption across four engineering tracks: optical design and simulation, electronics and PCB design, structural/CAD/CAE workflows, and FPGA design. Entries are organized around real tasks such as lens optimization, schematic and PCB review, routing, CAD generation, stress and topology analysis, drawing inspection, RTL generation, verification, synthesis, timing closure, and place-and-route. The collection emphasizes 2020+ research and usable open source components, including deterministic tools that agents can call for execution, simulation, verification, optimization, and artifact delivery.

中文

面向实体产品研发的 AI 智能体、基础模型、论文与开源工程工具持续更新清单， 重点跟踪四类工程岗位中的 AI 渗透：光学设计与仿真、电子与 PCB 设计、结构/CAD/CAE 工作流，以及 FPGA 设计。条目按照真实任务组织，覆盖镜头优化、原理图与 PCB 审查、 自动布线、CAD 生成、受力与拓扑分析、图纸检查、RTL 生成、验证、综合、时序收敛和 布局布线等。清单优先收录 2020 年以来的研究和可使用的开源组件，也收录能够被智能体 调用、用于执行、仿真、验证、优化和交付工程成果的确定性工具。

Why This List

Foundation models have already made rapid progress in text generation, image and video creation, and software development. General-purpose coding agents and work assistants such as Codex, OpenHands, and WorkBuddy can now complete or substantially accelerate many workflows in these areas. Their impact is highly visible in software and digital-content jobs, but adoption is less uniform in engineering roles that depend on domain physics, specialized EDA/CAD tools, proprietary data, formal verification, manufacturing constraints, and responsibility for physical products.

This repository looks at that gap from the perspective of an intelligent hardware company. It focuses on four roles where foundation models and agents are still moving from research prototypes into practical workflows: optical engineering, electronics and hardware design, structural and CAD engineering, and FPGA engineering. The list covers work such as lens design, optical simulation, schematic generation, PCB placement and routing, drawing review, structural and stress analysis, topology and wall-thickness optimization, RTL generation, verification, HLS, synthesis, timing closure, and FPGA place-and-route.

Our goal is not only to collect resources, but also to track how AI penetrates these engineering positions over time: which tasks have reliable agents and open implementations, which remain research problems, and where human engineering judgment is still essential. The collection therefore prioritizes 2020+ papers and actively maintained open-source projects, organized by concrete engineering workflow rather than by model family.

Note

An open-source project must publish an explicit license. Paper code or datasets without a declared license are listed separately and are not called open source. Always check bundled models, datasets, and vendor EULAs.

Important

This is an engineering-agent stack, not an AI-model-only list. Deterministic CAD/EDA tools, optical simulators, solvers, linters, formal engines, synthesizers, and place-and-route systems are in scope whenever an agent can invoke them to create, analyze, simulate, verify, optimize, or deliver an engineering artifact. Native AI/agent projects and conventional toolchain components are listed separately so their roles remain clear.

Canonical upstream repositories are preferred over forks and mirrors. A fork is listed separately only when it contains material, maintained changes that are relevant to engineering agents.

Contents

• Optical Engineering
• Hardware and PCB Design
• Structural and CAD Engineering
• FPGA Engineering
• Contributing
• License

Optical Engineering

Design, Simulation, and Inverse Design Papers

Year	Venue	Paper	Workflow
2026	arXiv	End-to-End Differentiable Design of Geometric Waveguide Displays	Waveguide-display simulation and optimization
2025	DATE	MAPS: Multi-Fidelity AI-Augmented Photonic Simulation and Inverse Design Infrastructure	Photonic simulation and inverse design
2024	Advanced Engineering Informatics	Artificial intelligence in optical lens design	Lens-design survey
2024	arXiv	TorchOptics: An open-source Python library for differentiable Fourier optics	Differentiable wave optics
2024	Nature Communications	Curriculum learning for ab initio deep learned refractive optics	Automatic lens design
2024	SIGGRAPH Asia	End-to-End Hybrid Refractive-Diffractive Lens Design with Differentiable Ray-Wave Model	Hybrid refractive-diffractive lens optimization
2024	Optics Express	End-to-End Automatic Lens Design with a Differentiable Diffraction Model	Diffraction-aware lens optimization
2023	ASPLOS	LightRidge: An End-to-end Agile Design Framework for Diffractive Optical Neural Networks	Diffractive optical neural networks
2022	TCI	dO: A Differentiable Engine for Deep Lens Design of Computational Imaging Systems	Differentiable geometric-optics engine
2021	TOG	End-to-End Complex Lens Design with Differentiable Ray Tracing	Learned compound-lens design
2020	TOG	Neural Holography with Camera-in-the-loop Training	Hologram generation and calibration
2020	CVPR	Deep Optics for Single-Shot High-Dynamic-Range Imaging	Learned optical coding

Agentic Optical Design Papers

Year	Paper	Workflow
2026	OptiAgent: End-to-End Optimization Modeling via Multi-Agent Iterative Refinement	Multi-agent optimization modeling
2026	OPTIAGENT: A Physics-Driven Agentic Framework for Automated Optical Design	Physics-validated optical design agent

Open-Source Projects

Project	Engineering use	License
DeepLens	Differentiable geometric optics and automatic lens design	Apache-2.0
DeepTrack 2.0	Microscopy, particle tracking, and optical characterization	MIT
diffractsim	Differentiable diffraction simulation	MPL-2.0
Dolphindes	Bounds and optimization for photonic inverse design	MIT
MAPS	Multi-fidelity AI photonic simulation and inverse design	MIT
pinn-shaper	PINN-based flat-optics beam shaping	MPL-2.0
TorchOptics	PyTorch Fourier-optics simulation and optimization	MIT
waveprop	Free-space propagation and holography models	MIT

Agent Toolchain Components

These libraries provide deterministic simulation, propagation, ray-tracing, and analysis capabilities that an optical agent can call and validate against.

Project	Agent role	Interface	License
HCIPy	Optical propagation and high-contrast imaging simulation	Python	MIT
POPPY	Physical-optics propagation and point-spread-function analysis	Python	BSD-3-Clause
prysm	Physical optics, phase retrieval, aberration analysis, and ray tracing	Python	MIT
ray-optics	Scriptable geometric-optics scene construction and ray simulation	Web / JavaScript	Apache-2.0

Hardware and PCB Design

This track covers IC/SoC physical design, analog layout, PCB schematic and routing, design-rule checking, thermal analysis, and drawing review. FPGA RTL generation and implementation are in FPGA Engineering.

Schematic, PCB, and CAD Generation Papers

Year	Venue	Paper	Workflow
2026	arXiv	SchGen: PCB Schematic Generation with Semantic-Grounded Code Representations	Schematic generation
2026	arXiv	PCBSchemaGen: Constraint-Guided Schematic Design via LLM	Constraint-guided schematic design
2026	ICLR	PCB-Bench: Benchmarking LLMs for Printed Circuit Board Placement and Routing	PCB placement and routing benchmark
2026	ESWA	Automation of PCB Autorouting via World-Model Reinforcement Learning and Freerouting Integration	Freerouting-grounded PCB routing loop
2024	IEEE	AI-Optimized Placement and Routing for PCB Design	PCB placement and routing
2024	MLCAD	OpenROAD-Assistant: An Open-Source Large Language Model for Physical Design Tasks	OpenROAD knowledge and script assistance
2024	MLCAD	An ML-aided Approach to Automatically Generate Schematic Symbols in PCB EDA Tools	PCB schematic-symbol generation
2024	ICCCAS	Reinforcement Learning-Based Placement Method for Printed Circuit Board	PCB component placement
2024	arXiv	LLM4EDA: Emerging Progress in Large Language Models for Electronic Design Automation	EDA survey
2024	arXiv	CadVLM: Bridging Language and Vision in the Generation of Parametric CAD	Drawing and CAD understanding
2023	TCAD	TRouter: Thermal-driven PCB Routing via Non-Local Crisscross Attention Networks	Thermal-aware routing
2023	ASP-DAC	DPRoute: Deep Learning Framework for Package Routing	Package and board routing
2023	ISPD	Goal Driven PCB Synthesis Using Machine Learning and CloudScale Compute	Constraint-driven PCB synthesis
2023	NeurIPS Workshop	CAD-LLM: Large Language Model for CAD Generation	Text-to-CAD
2022	DAC	Automated Accelerator Optimization Aided by Graph Neural Networks	Hardware design-space exploration
2022	ML4CAD	A Thermal Machine Learning Solver for Chip Simulation	Thermal simulation
2021	NeurIPS	On Joint Learning for Solving Placement and Routing in Chip Design	Joint placement and routing
2021	DAC	AutoSVA: Democratizing Formal Verification of RTL Module Interactions	Design review and formal checks
2020	ICCAD	A Customized Graph Neural Network Model for Guiding Analog IC Placement	Analog placement

Agentic EDA and Hardware Papers

Year	Paper	Workflow
2026	ZhuLong: Execution-Grounded LLM Agent for EDA Scripting with Offline API Self-Exploration	EDA scripting agent
2026	EDATracer: An Agentic Framework for Large-Scale EDA Artifact Analysis	EDA artifact review
2026	Can AI Agents Really Complete RTL-to-GDS? Lessons from Benchmarking Tool-Interactive EDA Workflows	End-to-end RTL-to-GDS agents
2026	SABLE: An NDA-Safe Closed-Loop LLM Framework for Analog Circuit Optimization in Industrial EDA Flows	Analog circuit optimization
2025	AutoEDA: Enabling EDA Flow Automation through Microservice-Based LLM Agents	EDA flow orchestration
2025	JARVIS: A Multi-Agent Code Assistant for High-Quality EDA Script Generation	EDA script generation
2024	Agentic-HLS: An Agentic Reasoning Based High-Level Synthesis System Using Large Language Models	HLS code and directive generation
2024	EDA-Aware RTL Generation with Large Language Models	RTL generation with synthesis feedback
2024	Automatically Improving LLM-based Verilog Generation using EDA Tool Feedback	Tool-grounded Verilog repair
2024	IICPilot: An Intelligent Integrated Circuit Backend Design Framework Using Open EDA	Backend design automation
2024	Ask-EDA: A Design Assistant Empowered by LLM, Hybrid RAG and Abbreviation De-hallucination	EDA design assistant
2023	ChatEDA: A Large Language Model Powered Autonomous Agent for EDA	Autonomous EDA assistant

Open-Source Projects

Project	Engineering use	License
ALIGN	Analog schematic-to-layout automation	BSD-3-Clause
CircuitLM	Multi-agent natural-language-to-schematic generation	MIT
DeepPCB KiCad plugin	AI-assisted PCB layout workflows in KiCad	Apache-2.0
DREAMPlace	GPU-accelerated placement optimization	BSD-3-Clause
KiCad	Schematic capture, PCB layout, DRC, and 3D inspection foundation	GPL-3.0
KiCad Copilot	LLM assistant for KiCad projects	MIT
KiCad MCP	Agent access to KiCad schematics, boards, footprints, and design checks	MIT
MAGICAL	Machine-learning-assisted analog layout	BSD-3-Clause
OpenLane	Automated RTL-to-GDS design flow	Apache-2.0
OpenROAD	Placement, routing, timing, and physical-design automation	BSD-3-Clause
PCBFlow	Programmatic PCB generation and layout	MIT
SchGen	Semantic-grounded-code PCB schematic generation	MIT

Agent Toolchain Components

Project	Agent role	Interface	License
eda-agent	Agent-controlled Altium, KiCad, and EasyEDA Pro schematic/PCB editing, review, audits, placement, and panelization	MCP / EDA APIs	Apache-2.0
freerouting	PCB autorouting backend for DSN/SES-based board workflows	CLI / GUI	GPL-3.0
KLayout	Scriptable GDS/OASIS layout viewing, editing, DRC, and extraction	Python / Ruby / GUI	GPL-3.0
LibrePCB	Schematic capture, PCB layout, library management, and manufacturing export	GUI / CLI	GPL-3.0
OpenSTA	Static timing analysis and timing-report feedback for chip-design agents	Tcl / CLI	GPL-3.0
SKiDL	Programmatic circuit and netlist generation with electrical-rule checks	Python	MIT

Datasets and Code with Unclear Licensing

• CircuitNet — EDA datasets for congestion, timing, power, and IR-drop prediction. The repository does not currently expose a standard software license; verify terms before reuse.
• PCBSchemaGen — research code for the 2026 schematic-generation paper; no explicit repository license was found.
• Pro-CAD — official code for proactive requirements clarification and text-to-CAD generation; no explicit repository license was found.

New PCB Agent and Multimodal Benchmark Papers

Year	Paper	Workflow
2026	OmniRouting: A Semantic-Coupled Multimodal Benchmark for Constraint-Aware Spatial Reasoning in PCB Routing	PCB routing reasoning
2026	PCBWorld: A Benchmark Environment for Engine-Grounded PCB Design Automation	Tool-interactive PCB agents
2026	OmniLayout: A Schematic-Coupled Multimodal Benchmark for Constraint-Aware Geometric Reasoning in PCB Layout	PCB layout reasoning
2026	PCB-QA: Evaluating LLMs over the First Printed Circuit Board Design Question-Answer Dataset	PCB design knowledge evaluation
2026	OmniSch: A Multimodal PCB Schematic Benchmark for Structured Diagram Visual Reasoning	Schematic review and reasoning
2026	HWE-Bench: Can Language Models Perform Board-level Schematic Designs?	Board-level schematic benchmark
2026	CircuitLM: A Multi-Agent LLM-Aided Design Framework for Generating Circuit Schematics from Natural Language Prompts	Multi-agent schematic generation
2025	EEschematic: Multimodal-LLM Based AI Agent for Schematic Generation of Analog Circuit	Analog schematic agent
2024	Schemato: An LLM for Netlist-to-Schematic Conversion	Netlist-to-schematic conversion
2024	AmpAgent: An LLM-based Multi-Agent System for Multi-stage Amplifier Schematic Design	Amplifier schematic design

Commercial EDA Signals

These proprietary systems are tracked as evidence of industry adoption, not as open-source recommendations. Vendor claims should be validated independently.

Product	Engineering workflow	Status
Cadence AuraStack AI Super Agent	PCB and advanced-package planning, constraints, implementation, DFM, and multiphysics orchestration	Announced
Siemens Fuse EDA AI Agent	Semiconductor, 3D-IC, and PCB design, verification, and sign-off workflows	Production
Synopsys AgentEngineer	RTL, lint, testbench, and verification workflows	Production

Structural and CAD Engineering

The structural track includes generative CAD, topology and thickness-aware design, load/stress prediction, finite-element surrogates, physics-informed simulation, and drawing/BIM review.

Agentic CAD and Structural Design Papers

Year	Paper	Workflow
2026	RA-CAD: Learning Post-Execution Critique for State-Aware Text-to-CAD Generation	State-aware CAD repair
2026	TraceCAD: Trace-Guided Repair for Agentic CAD Generation	Execution-trace CAD repair
2026	CADIR: A Cross-Backend Editable Intermediate Representation for Agentic CAD Generation	Cross-CAD-backend generation
2026	ArtisanCAD: An Industrial-Level CAD Agent with Expert-Grounded Knowledge Distillation	Industrial CAD agent
2026	ASSEMCAD: Production-Ready CAD Assembly Generation from Natural Language	Assembly generation with mating constraints
2026	AgentsCAD: Automated Design for Manufacturing of FDM Parts via Multi-Agent LLM Reasoning and Geometric Feature Recognition	Design for additive manufacturing
2026	Embodied CAD: Solver-Grounded LLM Agents for Parametric B-Rep Assembly Modeling	Parametric assembly modeling
2026	IterCAD: An Iterative Multimodal Agent for Visually-Grounded CAD Generation and Editing	CAD generation and editing
2026	Physics-in-the-Loop: A Hybrid Agentic Architecture for Validated CAD Engineering Design	Physics-validated CAD design
2026	Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback	FEA-in-the-loop CAD optimization
2026	Zero-to-CAD: Agentic Synthesis of Interpretable CAD Programs at Million-Scale Without Real Data	Large-scale CAD program synthesis
2026	ArtiCAD: Articulated CAD Assembly Design via Multi-Agent Code Generation	Articulated assembly design
2026	TOOLCAD: Exploring Tool-Using Large Language Models in Text-to-CAD Generation with Reinforcement Learning	Tool-using text-to-CAD agent
2026	CADSmith: Multi-Agent CAD Generation with Programmatic Geometric Validation	Geometry-validated CAD generation
2026	Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation	Interactive requirements clarification
2025	CADDesigner: Conceptual CAD Model Generation with a General-Purpose Agent	Conceptual CAD generation
2025	cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning	CAD reconstruction from images, point clouds, and text
2025	CAD-Llama: Leveraging Large Language Models for Computer-Aided Design Parametric 3D Model Generation	Parametric CAD command generation
2025	From Idea to CAD: A Language Model-Driven Multi-Agent System for Collaborative Design	Collaborative CAD design
2024	CAD-Assistant: Tool-Augmented VLLMs as Generic CAD Task Solvers	General CAD task assistant
2024	Text2CAD: Generating Sequential CAD Models from Beginner-to-Expert Level Text Prompts	Text-to-parametric-CAD generation

Parametric CAD Generation and Reconstruction Papers

Year	Paper	Workflow
2026	CADENA: Stepwise CAD Reverse Engineering	Stepwise mesh-to-parametric-CAD reconstruction
2026	HierCAD: Hierarchical Text-to-CAD Design via Structure Alignment and Parameter Grounding	Hierarchical text-to-CAD generation
2026	SOV-CAD: Stepwise Orthographic Views Guided CAD Modeling Sequence Reconstruction	Orthographic-view-to-CAD-sequence reconstruction
2026	UniCAD: A Unified Benchmark and Universal Model for Multi-Modal Multi-Task CAD	Multimodal CAD generation and understanding
2026	CADFit: Precise Mesh-to-CAD Program Generation with Hybrid Optimization	Editable CAD program recovery from meshes
2026	CADReasoner: Iterative Program Editing for CAD Reverse Engineering	Iterative CAD program reconstruction and repair
2026	GIFT: Bootstrapping Image-to-CAD Program Synthesis via Geometric Feedback	Geometry-grounded image-to-CAD synthesis
2026	Towards High-Fidelity CAD Generation via LLM-Driven Program Generation and Text-Based B-Rep Primitive Grounding	Text-to-CAD with B-Rep primitive grounding
2026	DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces	Editable B-Rep generation from multimodal input
2026	STEP-LLM: Generating CAD STEP Models from Natural Language with Large Language Models	Natural-language-to-STEP generation
2025	From Intent to Execution: Multimodal Chain-of-Thought Reinforcement Learning for Precise CAD Code Generation	Precise multimodal CAD code generation
2025	CAD-Coder: An Open-Source Vision-Language Model for Computer-Aided Design Code Generation	Image-to-CAD code generation
2025	Text-to-CadQuery: A New Paradigm for CAD Generation with Scalable Large Model Capabilities	Text-to-CadQuery program generation
2024	TransCAD: A Hierarchical Transformer for CAD Sequence Inference from Point Clouds	Point-cloud-to-CAD-sequence reconstruction
2021	DeepCAD: A Deep Generative Network for Computer-Aided Design Models	CAD command-sequence generation baseline

Papers

Year	Venue	Paper	Workflow
2026	arXiv	Developing Large Language Model for BIM-based Design with Domain Knowledge	BIM design assistant
2025	arXiv	Optimize Any Topology: A Foundation Model for Shape- and Resolution-Agnostic Topology Optimization	Foundation model for topology optimization
2025	arXiv	Generative Topology Optimization: Exploring Diverse Solutions in Engineering Design	Diverse generative design
2025	NeurIPS	GeoCAD: Local Geometry-Controllable CAD Generation with Large Language Models	Parametric CAD generation
2024	arXiv	Structural Design Through Reinforcement Learning	Structural member/layout design
2024	arXiv	Multi-scale Topology Optimization using Neural Networks	Multi-scale topology optimization
2024	arXiv	A Dual Physics-Informed Neural Network for Topology Optimization	Physics-constrained optimization
2024	arXiv	NITO: Neural Implicit Fields for Resolution-free Topology Optimization	Resolution-free topology
2023	FEAD	A Finite-Element-Informed Neural Network for Parametric Simulation in Structural Mechanics	Load and response surrogate
2023	NeurIPS Workshop	TopologyGAN: Topology Optimization Using Generative Adversarial Networks	Generative topology optimization
2022	SMO	Deep Learning Accelerated Topology Optimization with Inherent Control of Image Quality	Fast topology generation
2022	SMO	Accelerated Topology Optimization Design of 3D Structures Based on Deep Learning	3D structural design
2021	Nature Machine Intelligence	Learning Nonlinear Operators via DeepONet	Simulation surrogate and operator learning

CAD Benchmarks and Reality Checks

Year	Benchmark	What it tests
2026	Text2CAD-Bench: A Benchmark for LLM-based Text-to-Parametric CAD Generation	Text-to-CAD generation from simple parts to complex topology
2026	MUSE: Benchmarking Manufacturable, Functional, and Assemblable Text-to-CAD Generation	Manufacturability, functionality, and assemblability of generated B-Rep parts
2026	CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation	CAD program generation across increasing geometric complexity
2026	BenchCAD: A Comprehensive, Industry-Standard Benchmark for Programmatic CAD	Executability and parametric structure of industrial-style CAD programs
2026	Text-to-CAD Evaluation with CADTests	Functional requirements tested directly on generated geometry

Visual similarity is insufficient for engineering CAD. Evaluation should also cover execution success, editable feature history, constraints, assemblies, manufacturability, and behavior under downstream analysis.

Agentic CAD Projects

Project	Engineering use	License
AgentSCAD	Generates and repairs OpenSCAD models with geometry and manufacturability validation	MIT
CADAM	Converts text or images into editable parametric OpenSCAD, STL, and DXF artifacts	GPL-3.0
CAD Skills	Agent skills for generating, inspecting, modifying, and validating CAD/CAE/CAM artifacts	MIT
FreeCAD AI	Conversational FreeCAD workbench that generates native geometry through Python	LGPL-2.1

Open-Source CAD Research Projects and Datasets

Project	Engineering use	License
CAD-Coder	Vision-language model and training code for image-to-CAD code generation	Apache-2.0
CADENA	Stepwise reverse engineering from meshes to parametric CAD programs	MIT
DeepCAD	CAD command-sequence generation baseline and dataset tooling	MIT
HierCAD	Hierarchical text-to-CAD generation with parameter grounding	MIT
SketchGraphs	Dataset of 15 million CAD sketches with geometric constraint graphs	MIT

CAD Research Code with Restricted or Unclear Licensing

• CADFit publishes research code under CC BY-NC 4.0, which restricts commercial use and is not treated here as an open-source software license.
• FutureCAD, SOV-CAD, and Text-to-CadQuery provide research code but currently expose no explicit repository license.

Open-Source Projects

Project	Engineering use	License
DeepXDE	PINNs and operator learning for PDE and mechanics problems	LGPL-2.1
DiffFE-Physics-Lab	Differentiable finite elements with ML integration	MIT
FEINN	Finite-element-informed neural networks for truss, beam, and plate response	Apache-2.0
IF-TONIR	Implicit neural representations for iteration-free topology optimization	GPL-2.0
NeuralOperator	Neural operators for fast field and response prediction	MIT
NVIDIA Warp	Differentiable GPU simulation and optimization for mechanics	Apache-2.0
PhysicsNeMo	Physics-informed surrogate models for structural mechanics and CFD	Apache-2.0
Text2BIM	Text-to-BIM generation and building-design prototyping	MIT
TopOpt.jl	Differentiable topology-optimization components	MIT

Additional AI/ML for Structural Simulation

Project	Engineering use	License
DL4TO	Deep learning and differentiable physics for 3D topology optimization and linear elasticity	Apache-2.0
NeuralSolver	Benchmark library for neural PDE solvers on regular and general geometries	MIT
PINNacle	Reproducible benchmark for physics-informed neural PDE solvers	MIT
The Well	Large-scale physics-simulation datasets for surrogate and foundation models	BSD-3-Clause

CAE and Simulation Agent Papers

Year	Paper	Workflow
2025	Foam-Agent 2.0: An End-to-End Composable Multi-Agent Framework for Automating CFD Simulation in OpenFOAM	Composable OpenFOAM workflow automation
2025	CFDagent: A Language-Guided, Zero-Shot Multi-Agent System for Complex Flow Simulation	Autonomous CFD case generation and execution
2025	ChatCFD: An LLM-Driven Agent for End-to-End CFD Automation with Structured Knowledge and Reasoning	End-to-end CFD workflow
2025	OpenFOAMGPT: a RAG-Augmented LLM Agent for OpenFOAM-Based Computational Fluid Dynamics	OpenFOAM knowledge and case assistance
2024	MetaOpenFOAM: an LLM-based multi-agent framework for CFD	Multi-agent OpenFOAM workflow

CAE Solver and Feedback-Loop Papers

These papers document recent deterministic solver capabilities that can serve as execution and validation backends for structural-engineering agents.

Year	Venue	Paper	Workflow
2024	IJHPCA	High-Performance Finite Elements with MFEM	Scalable finite-element execution backend
2023	CMC	Fundamental Study on Fracture Analysis of Concrete Materials Using OpenRadioss	OpenRadioss concrete-fracture simulation
2022	IJHPCA	Matrix-Free Approaches for GPU Acceleration of a High-Order Finite Element Hydrodynamics Application Using MFEM, Umpire, and RAJA	GPU-accelerated simulation feedback
2021	CAMWA	MFEM: A Modular Finite Element Methods Library	Modular structural and multiphysics solving

Open-Source CAE Agents

Project	Engineering use	License
AutoCFD	Converts natural-language requirements into executable OpenFOAM cases with iterative correction	GPL-3.0
MetaOpenFOAM	Multi-agent OpenFOAM planning, case generation, execution, and result analysis	GPL-3.0

Agent-Ready CAE Infrastructure

These projects are not necessarily AI models themselves. They are included because their Python, CLI, or MCP interfaces make them practical solver, geometry, meshing, and post-processing backends for engineering agents.

Project	Engineering use	Interface	License
build123d	Scriptable parametric solid modeling for agent-generated CAD	Python	Apache-2.0
CadQuery	Parametric CAD generation and STEP/STL export	Python	Apache-2.0
DOLFINx	Parallel finite-element analysis and PDE solving	Python / C++	LGPL-3.0
freecad-mcp	Agent-controlled FreeCAD modeling, inspection, Python, and FEM workflows	MCP / Python	MIT
Netgen	Automatic tetrahedral meshing from STEP, IGES, STL, and CSG	Python / CLI	LGPL-2.1
ParaView-MCP	Agent-controlled simulation visualization and post-processing	MCP / Python	BSD-3-Clause
SfePy	Scriptable structural, thermal, and multiphysics FEM	Python	BSD-3-Clause
sim-cli	Replayable agent runtime for COMSOL, Abaqus, and Ansys simulation workflows	CLI / plugins	Apache-2.0
viznoir	Headless MCP visualization for VTK, CGNS, and OpenFOAM results	MCP / Python	MIT

Additional CAD, Meshing, and Solver Components

Project	Agent role	Interface	License
CalculiX	Linear/nonlinear structural, thermal, and dynamic finite-element analysis	CLI / files	GPL-2.0
FreeCAD	Parametric solid modeling, drawing, assembly, and FEM automation	Python / GUI	LGPL-2.1
MFEM	Scalable finite-element backend for structural and multiphysics workflows	C++ / Python	BSD-3-Clause
meshio	Conversion between common finite-element and mesh formats	Python / CLI	MIT
OpenFOAM	CFD and multiphysics case generation, execution, and result production	CLI / files	GPL-3.0
OpenRadioss	Explicit finite-element solver for impact, crash, drop, and dynamic-event analysis	CLI / files	AGPL-3.0
PyVista	Mesh analysis, field inspection, and 3D simulation-result validation	Python	MIT
trimesh	Mesh loading, repair, measurements, collision, and geometric validation	Python	MIT

The agent-callability perspective and several infrastructure candidates in this section were cross-checked against awesome-ai-cae, which maintains a broader ranked collection of open CAE, CAD, meshing, and visualization tools.

The AI-versus-foundation distinction, maturity-oriented curation, and several CAD-agent, simulation-agent, benchmark, and industry-signal candidates were also cross-checked against Awesome-Physical-Engineering-AI.

The parametric-CAD generation, reconstruction, dataset, and evaluation entries were additionally cross-checked against the bilingual Awesome AI4CAD Hub, whose broader automated index is useful for discovery. Entries here were verified against their original papers and repositories because the hub also contains duplicates, pre-2020 work, and incomplete metadata.

Commercial Drawing Review and Engineering Knowledge Signals

These products are proprietary and are listed only to track adoption in tasks for which reproducible open implementations remain scarce.

Product	Engineering workflow	Status
bananaz	CAD, drawing, BOM, change, standards, and DFX review	Production
CoLab AutoReview	Automated dimensions, tolerances, standards, BOM, and DFM review	Production
DraftAid	Generation of fabrication drawings from 3D CAD	Production
Leo AI	Retrieval and reasoning over PLM, PDM, CAD, and technical data	Production

Wall-thickness checks, load-path validation, tolerance review, and drawing approval are emerging industrial workflows. Contributions with a reproducible paper, dataset, plugin, or benchmark are especially useful here.

FPGA Engineering

RTL generation, simulation, assertion checking, HLS, quantization, synthesis, place-and-route, timing prediction, and bitstream-oriented agent workflows.

RTL, Verification, and HLS Papers

Year	Venue	Paper	Workflow
2026	ASP-DAC	LLM-Assisted Circuit Verification: A Comprehensive Survey	Verification survey
2025	arXiv	REvolution: An Evolutionary Framework for RTL Generation	Iterative RTL generation
2025	TODAES	High-level Synthesis Directives Design Optimization via Large Language Model	HLS directive optimization
2025	ASP-DAC	LLSM: LLM-enhanced Logic Synthesis Model	Logic synthesis
2025	ASP-DAC	AssertLLM: Generating Hardware Verification Assertions from Design Specifications	Assertion generation
2025	arXiv	OpenLLM-RTL: Open Dataset and Benchmark for LLM-Aided Design RTL Generation	RTL benchmark
2025	ACM	ResBench: A Resource-Aware Benchmark for LLM-Generated FPGA Designs	FPGA resource benchmark
2024	TCAD	RTLCoder: Fully Open-Source and Efficient LLM-Assisted RTL Code Generation	RTL generation
2024	ICCAD	OpenLLM-RTL: Open Dataset and Benchmark for LLM-Aided Design RTL Generation	RTL evaluation
2024	ASP-DAC	RTLLM: An Open-Source Benchmark for Design RTL Generation with Large Language Models	RTL benchmark
2024	TODAES	Survey of Machine Learning for Software-Assisted Hardware Design Verification	Verification survey
2024	arXiv	Revisiting VerilogEval: Newer LLMs, In-Context Learning, and Specification-to-RTL Tasks	Spec-to-RTL evaluation
2023	ICCAD	VerilogEval: Evaluating Large Language Models for Verilog Code Generation	Verilog benchmark
2023	ICCAD	Robust GNN-based Representation Learning for HLS	HLS prediction
2022	DAC	High-Level Synthesis Performance Prediction Using GNNs	QoR prediction
2021	JMLR	hls4ml: An Open-Source Codesign Workflow for Low-Power ML Devices	ML-to-HLS
2020	FPL	LogicNets: Co-Designed Neural Networks and Circuits for Extreme-Throughput Applications	LUT-based inference

New FPGA and RTL Agent Papers

Year	Paper	Workflow
2026	HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translation	C/C++ to HLS translation
2026	VPR-Evolve: Multi-Agent-Driven Algorithm Evolution for FPGA Place and Route	FPGA place-and-route optimization
2026	CHIA: An Open-Source Framework for Principled, Agentic AI-Driven Hardware/Software Co-design Research	Hardware/software co-design agents
2026	HSCO-Bench: An Agent-Driven End-to-End Hardware-Software Co-design Benchmark for Systems-on-Chip	SoC co-design benchmark
2026	Design Conductor 2.0: An Agent Builds a TurboQuant Inference Accelerator in 80 Hours	Accelerator design agent
2026	ChipCraftBrain: Validation-First RTL Generation via Multi-Agent Orchestration	Multi-agent RTL generation
2026	Dr. RTL: Autonomous Agentic RTL Optimization through Tool-Grounded Self-Improvement	RTL optimization agent
2026	VeriAgent: A Tool-Integrated Multi-Agent System with Evolving Memory for PPA-Aware RTL Code Generation	PPA-aware RTL generation
2026	LAAFD: LLM-based Agents for Accelerated FPGA Design	FPGA design automation
2025	A2H-MAS: An Algorithm-to-HLS Multi-Agent System for Automated and Reliable FPGA Implementation	Algorithm-to-HLS implementation
2025	TimelyHLS: LLM-Based Timing-Aware and Architecture-Specific FPGA HLS Optimization	Timing-aware HLS optimization

FPGA Toolchain and Closed-Loop Papers

These system and workflow papers cover the parsers, simulators, architecture models, and place-and-route engines that provide executable feedback to FPGA agents.

Year	Venue	Paper	Workflow
2025	TRETS	VTR 9: Open-Source CAD for Fabric and Beyond FPGA Architecture Exploration	FPGA architecture exploration and routing
2025	FPL	Cocotb-Pynq: Co-Simulating Python+RTL Applications Targeting Pynq Platforms with Cocotb	Python/RTL co-simulation
2024	ISEDA	OpenPARF 3.0: Robust Multi-Electrostatics Based FPGA Macro Placement Considering Cascaded Macros Groups and Fence Regions	Constraint-aware FPGA macro placement
2024	EIECT	Cocotb-Based Verification of Multi-Protocol Interconnection Network	Cocotb-driven functional verification
2023	ASICON	OpenPARF: An Open-Source Placement and Routing Framework for Large-Scale Heterogeneous FPGAs with Deep Learning Toolkit	ML-ready FPGA placement and routing
2022	SBCCI	Integrating Machine-Learning Probes into the VTR FPGA Design Flow	ML instrumentation in VTR
2020	WOSET	Universal Hardware Data Model	Surelog/UHDM interoperability for open EDA tools

Open-Source Projects

Project	Engineering use	License
Brevitas	Quantization-aware training for FPGA deployment	BSD-3-Clause
CIRCT	MLIR-based RTL and hardware compilation infrastructure	Apache-2.0 WITH LLVM exception
FINN	Dataflow compilation for quantized neural networks	BSD-3-Clause
F4PGA	Open FPGA synthesis and place-and-route flows	Apache-2.0
FPGA-Agent-skills	Guided HLS, simulation, synthesis, timing, and debug skills	GPL-2.0
hls4ml	Converts trained models to HLS implementations	Apache-2.0
nextpnr	FPGA place-and-route backend used by open flows	ISC
SiliconCompiler	Portable synthesis, simulation, and FPGA APR orchestration	Apache-2.0
TAPA	Task-parallel HLS to RTL and software/RTL co-simulation	MIT
Verilator	Fast SystemVerilog simulation and linting for generated RTL	LGPL-3.0
VerilogEval	Reproducible LLM-to-Verilog evaluation harness	MIT
Vitis AI	Quantization, compilation, runtime, and deployment on AMD devices	Apache-2.0
Yosys	RTL synthesis and formal tooling for open FPGA flows	ISC

Agent Toolchain Components

These tools can form the executable feedback loop for an FPGA agent, from RTL parsing and lint through simulation, formal verification, synthesis, timing, and architecture-aware placement and routing.

Project	Agent role	Interface	License
cocotb	Python testbench generation, simulation control, and regression checking	Python / VPI	BSD-3-Clause
GHDL	VHDL analysis, elaboration, simulation, and synthesis integration	CLI	GPL-2.0
Icarus Verilog	Verilog/SystemVerilog compilation and simulation	CLI	GPL-2.0
OpenPARF	Large-scale heterogeneous FPGA placement and routing with a deep-learning toolkit	Python / C++	BSD-3-Clause
Surelog	SystemVerilog preprocessing, parsing, elaboration, UHDM, and AST access	CLI / C++ / Python	Apache-2.0
SymbiYosys	Orchestration of Yosys-based bounded, unbounded, and cover formal checks	CLI	ISC
Verible	SystemVerilog parsing, linting, formatting, and language-server checks	CLI / LSP	Apache-2.0
VTR	FPGA architecture exploration and Verilog-to-routing flow with VPR	CLI	Mixed, primarily MIT

Research Code without an Explicit License

• RTLCoder — official code and datasets for the RTLCoder/OpenLLM-RTL papers; no repository license is currently declared.

Contributing

Read CONTRIBUTING.md (简体中文) before opening a pull request.

1. Add only 2020+ papers, or projects with meaningful 2020+ maintenance.
2. Put each item in one primary role and one concrete workflow subsection.
3. Prefer official proceedings, DOI, arXiv, and canonical repositories.
4. List a project in the open-source table only when its repository exposes an explicit license; otherwise use the research-code note.
5. Keep papers newest first and projects alphabetically sorted within each table.
6. Include the engineering artifact being generated, analyzed, checked, or optimized (for example RTL, PCB routing, stress field, or lens parameters).

License

This list is released under the MIT License. Linked papers, code, datasets, models, and vendor tools remain under their own licenses.