pith. sign in

arxiv: 2606.00097 · v3 · pith:TTGQS7SGnew · submitted 2026-05-25 · 💻 cs.RO · cs.MA

RocketSmith: Agentic Additive Manufacturing of High-Powered Rockets

Peter Pak , Jesse Barkley , Rumi Loghmani , Derek Baich , Ananya Pamal , Amir Barati Farimani This is my paper

Pith reviewed 2026-06-30 11:41 UTC · model grok-4.3

classification 💻 cs.RO cs.MA
keywords agentic systemsadditive manufacturingrocket designlarge language modelsDFAMflight testinghigh power rocketsparametric design
0
0 comments X

The pith

An LLM-orchestrated agentic system automates rocket design for additive manufacturing and produces four stable-flying high-power rockets that reach 80% of predicted altitude.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

RocketSmith uses a large language model to coordinate software tools that check rocket flight stability and create parametric designs for 3D printing. The system ran optimization workflows in both fully automated and assisted modes to produce four different high-powered rockets. These were printed, inspected, and flown at a launch event. Every rocket flew stably, two were recovered intact for future use, and flight data showed they reached 80 percent of the height the system had predicted.

Core claim

RocketSmith is an agentic system which intelligently automates the DFAM process for the development of high powered rockets suitable for launch. The system utilizes a large language model to orchestrate the execution of software tools to validate design characteristics such as flight stability and generate the parametric design components for the rocket assembly. With this system, four distinct high power rockets with various motor and assembly configurations were developed utilizing the unique design capabilities of additive manufacturing. These assembly components were fabricated using various FDM printers, manually evaluated for flight readiness, and flight tested at a launch event. From

What carries the argument

The LLM-orchestrated agentic system with subagents and skills that enable iterative optimization of flight parameters in zero-shot and human-in-the-loop workflows.

If this is right

  • Four distinct rockets with different motor and assembly configurations can be developed and fabricated using additive manufacturing.
  • All rockets achieve stable launch during testing.
  • Two of the four rockets can be successfully recovered in reflyable condition.
  • Altimeter data confirms the rockets reach an altitude of 80% of the apogee predicted by the system.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This automation could reduce the time and expertise needed to iterate on rocket designs for hobbyist or small-team projects.
  • Closing the gap between predicted and actual performance might require tighter coupling to higher-fidelity physics simulators.
  • Similar agentic workflows may apply to designing other additively manufactured vehicles that require stability and performance validation.

Load-bearing premise

The simulation and validation tools orchestrated by the LLM produce sufficiently accurate predictions of real-world flight stability and apogee.

What would settle it

A follow-up launch campaign in which additional rockets designed by the same system either fail to achieve stable flight or reach substantially less than 80% of the predicted apogee would show the automation does not reliably produce functional designs.

Figures

Figures reproduced from arXiv: 2606.00097 by Amir Barati Farimani, Ananya Pamal, Derek Baich, Jesse Barkley, Peter Pak, Rumi Loghmani.

Figure 1
Figure 1. Figure 1: RocketSmith utilized as a Claude Code 4 plugin enables the efficient development of high powered rockets. The agentic system is capable of designing an OpenRocket 5 based component tree with provided user constraints such as specific motors and flight character￾istics. Prescribed dimensions from the component tree are utilized to generate parametric models of the rocket airframe using CADSmith. 6 More prec… view at source ↗
Figure 2
Figure 2. Figure 2: (Left to Right) High Power 1 launched by Pak with an AeroTech H100W motor, [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: (Left) OpenRocket 5 provides the foundational toolkit for generating rocket designs and running flight simulations. (Right) RocketSmith establishes a component tree to use for the generation of STEP file components and design related skills and subagents. 4.1.2 build123d build123d 44 is a parametric CAD Python library that wraps the OpenCASCADE31 geo￾metric kernel and uses context-manager and operator-over… view at source ↗
Figure 4
Figure 4. Figure 4: (Left) build123d is the primary framework to use for generating the parametric part configurations written in Python. (Right) STEP file is generated from running Python snippet and visualization allows for easy part adjustment. 15 [PITH_FULL_IMAGE:figures/full_fig_p015_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Weight estimation for generated STEP files for various components are obtained using Prusaslicer7 using configurations for the expected material. 4.2 Agentic System The development of high powered rockets is augmented with the usage of an agentic system which is capable of operating in zero-shot 47,48 and human-in-the-loop49 conditions. Under both operating conditions the system provides a graphical user i… view at source ↗
Figure 6
Figure 6. Figure 6: (Top Window) Terminal with Claude Code provides the primary means to con [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: (Left) Ender 3 with printed middle airframe component for High Power 2. (Middle) [PITH_FULL_IMAGE:figures/full_fig_p020_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Rocket design schematics for High Power 1 (Pak), 2 (Loghmani), 3 (Barkley), and [PITH_FULL_IMAGE:figures/full_fig_p021_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: (Left) Finished lower airframe printed using ABS filament over the course of several [PITH_FULL_IMAGE:figures/full_fig_p023_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: (Left) 38 mm motor tube inserted into the lower airframe to ensure adequate fit [PITH_FULL_IMAGE:figures/full_fig_p024_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: (Left) Lower airframe printed using ABS filament with motor tube inserted [PITH_FULL_IMAGE:figures/full_fig_p025_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: (Left) Finished lower airframe component printed with [PITH_FULL_IMAGE:figures/full_fig_p026_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: (Top Left) Recovered High Power 1 components after fracture from ejection [PITH_FULL_IMAGE:figures/full_fig_p028_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Recovered components of High Power 1 after successful launch including: (Left to [PITH_FULL_IMAGE:figures/full_fig_p029_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Recovered High Power 3 rocket after removal from tree with lower airframe split [PITH_FULL_IMAGE:figures/full_fig_p030_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Recovered High Power 4 rocket in reflyable condition with minor cosmetic blem [PITH_FULL_IMAGE:figures/full_fig_p031_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Flight data collected from StratoLogger altimeters on High Power 1 and High [PITH_FULL_IMAGE:figures/full_fig_p033_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: (Left) Voron-2-Tall during assembly, upside down to access internal electronics underneath the build plate. (Middle) First successful Benchy print with fully assembled printer and StealthBurner 59 toolhead using PLA filament. (Right) Voron-2-Tall with fin￾ished print for lower airframe component of Big Heavy 2 60 using ABS filament. 37 [PITH_FULL_IMAGE:figures/full_fig_p037_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: View of the launch site taken from High Power 1 onboard RunCam 5 camera just [PITH_FULL_IMAGE:figures/full_fig_p038_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Photo taken using RunCam 5 from High Power 4 around measured apogee of 570 [PITH_FULL_IMAGE:figures/full_fig_p039_20.png] view at source ↗
read the original abstract

RocketSmith is an agentic system which intelligently automates the DFAM process for the development of high powered rockets suitable for launch. The system utilizes a large language model to orchestrate the execution of software tools to validate design characteristics such as flight stability and generate the parametric design components for the rocket assembly. A collection of subagents and skills enable optimization workflows of flight parameters via iteration in both zero-shot and human-in-the-loop workflows. With this system, four distinct high power rockets with various motor and assembly configurations were developed utilizing the unique design capabilities of additive manufacturing. These assembly components were fabricated using various FDM printers, manually evaluated for flight readiness, and flight tested at a launch event. From these tests, all rockets achieved a stable launch and two of the four rockets were successfully recovered in reflyable condition. The altimeter data validated that the rockets achieved an altitude 80% of the expected apogee predicted by the agentic system, establishing consistency between simulation and experimentation.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces RocketSmith, an agentic system that uses a large language model to orchestrate software tools for validating rocket flight stability and generating parametric designs for additive manufacturing. Four distinct high-power rockets with varying motor and assembly configurations were developed, fabricated via FDM printing, manually checked, and flight-tested. All achieved stable launches, two were recovered in reflyable condition, and altimeter data showed altitudes at 80% of the agentic system's predicted apogee, which the abstract presents as establishing consistency between simulation and experiment.

Significance. If the results hold after addressing validation gaps, the work would demonstrate the viability of LLM-driven agentic workflows for end-to-end automation of complex aerospace design tasks that combine simulation, optimization, and physical hardware realization via additive manufacturing. The execution of actual flight tests at a launch event supplies empirical grounding that is uncommon in purely computational agentic studies.

major comments (2)
  1. [Abstract] Abstract: the claim that altimeter data 'validated' consistency with the agentic predictions is undermined by the reported 80% apogee achievement; no error bars, sensitivity analysis, or discussion of the 20% shortfall is supplied, leaving open whether the discrepancy arises from unmodeled effects, overstated thrust curves, or underestimated drag in the orchestrated simulations.
  2. [Abstract] Abstract: the central attribution of success to the agentic validation loop cannot be assessed because the manuscript supplies no description of the specific simulation packages used, input assumptions (Cd, motor thrust profiles, atmospheric models), or uncertainty quantification; without these the 80% figure cannot be interpreted as evidence of reliable prediction rather than systematic overprediction.
minor comments (2)
  1. The manuscript would benefit from explicit listing of the sub-agents, skills, and external tools invoked by the LLM, together with example prompts or workflow diagrams, to allow replication of the zero-shot and human-in-the-loop optimization loops.
  2. Flight-test results would be strengthened by inclusion of time-series altimeter traces, stability margin calculations, and a table comparing predicted versus measured apogee, mass, and motor performance for each of the four configurations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for highlighting important issues in the abstract regarding the strength of the simulation-experiment comparison. We agree that the current phrasing and lack of supporting details limit interpretability, and we will revise the manuscript accordingly to provide a more balanced presentation. Point-by-point responses to the major comments are below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that altimeter data 'validated' consistency with the agentic predictions is undermined by the reported 80% apogee achievement; no error bars, sensitivity analysis, or discussion of the 20% shortfall is supplied, leaving open whether the discrepancy arises from unmodeled effects, overstated thrust curves, or underestimated drag in the orchestrated simulations.

    Authors: We agree that the abstract's use of 'validated' is too strong given the 20% shortfall and that the manuscript lacks quantitative support for the consistency claim. In revision we will (1) change the abstract language to 'showed altitudes reaching approximately 80% of the predicted apogee, indicating reasonable agreement between simulation and flight data,' (2) add error bars derived from the altimeter records of the two recovered flights, (3) include a short sensitivity discussion in the results section examining the effects of plausible variations in drag coefficient and motor thrust, and (4) explicitly note possible sources of the shortfall such as real-world motor performance scatter and unmodeled launch conditions. These changes will be made without altering the reported flight outcomes. revision: yes

  2. Referee: [Abstract] Abstract: the central attribution of success to the agentic validation loop cannot be assessed because the manuscript supplies no description of the specific simulation packages used, input assumptions (Cd, motor thrust profiles, atmospheric models), or uncertainty quantification; without these the 80% figure cannot be interpreted as evidence of reliable prediction rather than systematic overprediction.

    Authors: We acknowledge that the manuscript does not currently provide the requested simulation details, which prevents readers from evaluating the reliability of the 80% figure. In the revised version we will expand the methods section with (1) the names and versions of the simulation packages orchestrated by the agent, (2) the specific input assumptions used (including the drag coefficient values, motor thrust curves taken from manufacturer data sheets, and the atmospheric model), and (3) any uncertainty quantification or Monte-Carlo-style checks performed by the agentic workflow. This added transparency will allow proper assessment of whether the observed shortfall reflects systematic bias or expected variability. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical flight results independent of simulation predictions

full rationale

The paper reports an LLM-orchestrated design workflow followed by physical fabrication and flight testing of four rockets. The key consistency claim rests on altimeter data showing 80% of simulated apogee, which is an external physical measurement rather than a quantity derived from or fitted to the same inputs. No equations, parameter fittings, self-citations, or uniqueness theorems are invoked that would reduce the reported outcomes to the system's own definitions or prior outputs by construction. The workflow uses external simulation tools and real-world tests, making the derivation chain self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides insufficient detail to enumerate specific free parameters, axioms, or invented entities; the system appears to rely on existing simulation tools and LLM capabilities without introducing new physical entities.

pith-pipeline@v0.9.1-grok · 5716 in / 1129 out tokens · 23216 ms · 2026-06-30T11:41:20.234848+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

70 extracted references · 42 canonical work pages

  1. [1]

    The International Journal of Advanced Manufacturing Technology , author =

    Additive manufacturing-enabled design theory and methodology: a critical review , volume =. The International Journal of Advanced Manufacturing Technology , author =. 2015 , keywords =. doi:10.1007/s00170-015-6994-5 , abstract =

    work page doi:10.1007/s00170-015-6994-5 2015

  2. [2]

    2008 , pages =

    Requirements for. 2008 , pages =

    2008

  3. [3]

    doi:10.5281/zenodo.14872323 , abstract =

    build123d:. doi:10.5281/zenodo.14872323 , abstract =

    work page doi:10.5281/zenodo.14872323

  4. [4]

    Yao, Shunyu and Zhao, Jeffrey and Yu, Dian and Du, Nan and Shafran, Izhak and Narasimhan, Karthik and Cao, Yuan , year =

  5. [5]

    and Wettig, Alexander and Lieret, Kilian and Yao, Shunyu and Narasimhan, Karthik and Press, Ofir , month = dec, year =

    Yang, John and Jimenez, Carlos E. and Wettig, Alexander and Lieret, Kilian and Yao, Shunyu and Narasimhan, Karthik and Press, Ofir , month = dec, year =. Proceedings of the 38th

  6. [6]

    and Le, Quoc V

    Wei, Jason and Bosma, Maarten and Zhao, Vincent and Guu, Kelvin and Yu, Adams Wei and Lester, Brian and Du, Nan and Dai, Andrew M. and Le, Quoc V. , month = oct, year =. Finetuned

  7. [7]

    repeated

    Continual. ACM Computing Surveys , author =. 2025 , pages =. doi:10.1145/3735633 , abstract =

    work page doi:10.1145/3735633 2025

  8. [8]

    and Burger, Doug and Wang, Chi , month = aug, year =

    Wu, Qingyun and Bansal, Gagan and Zhang, Jieyu and Wu, Yiran and Li, Beibin and Zhu, Erkang and Jiang, Li and Zhang, Xiaoyun and Zhang, Shaokun and Liu, Jiale and Awadallah, Ahmed Hassan and White, Ryen W. and Burger, Doug and Wang, Chi , month = aug, year =

  9. [9]

    and Le, Quoc V

    Wei, Jason and Wang, Xuezhi and Schuurmans, Dale and Bosma, Maarten and Ichter, Brian and Xia, Fei and Chi, Ed H. and Le, Quoc V. and Zhou, Denny , month = nov, year =. Chain-of-thought prompting elicits reasoning in large language models , isbn =. Proceedings of the 36th

  10. [10]

    2024 , keywords =

    Additive Manufacturing , author =. 2024 , keywords =. doi:10.1016/j.addma.2024.104503 , abstract =

    work page doi:10.1016/j.addma.2024.104503 2024

  11. [11]

    Additive Manufacturing Letters , author =

    Deep learning for melt pool depth contour prediction from surface thermal images via vision transformers , volume =. Additive Manufacturing Letters , author =. 2024 , keywords =. doi:10.1016/j.addlet.2024.100243 , abstract =

    work page doi:10.1016/j.addlet.2024.100243 2024

  12. [12]

    and Chandrasekhar, Achuth and Barati Farimani, Amir , month = feb, year =

    Ock, Janghoon and Meda, Radheesh Sharma and Badrinarayanan, Srivathsan and Aluru, Neha S. and Chandrasekhar, Achuth and Barati Farimani, Amir , month = feb, year =. Large. Journal of Chemical Information and Modeling , publisher =. doi:10.1021/acs.jcim.5c02454 , abstract =

    work page doi:10.1021/acs.jcim.5c02454

  13. [13]

    Learn to explain: multimodal reasoning via thought chains for science question answering , isbn =

    Lu, Pan and Mishra, Swaroop and Xia, Tony and Qiu, Liang and Chang, Kai-Wei and Zhu, Song-Chun and Tafjord, Oyvind and Clark, Peter and Kalyan, Ashwin , month = nov, year =. Learn to explain: multimodal reasoning via thought chains for science question answering , isbn =. Proceedings of the 36th

  14. [14]

    Large language models are zero-shot reasoners , isbn =

    Kojima, Takeshi and Gu, Shixiang Shane and Reid, Machel and Matsuo, Yutaka and Iwasawa, Yusuke , month = nov, year =. Large language models are zero-shot reasoners , isbn =. Proceedings of the 36th

  15. [15]

    Proceedings of the 19th

    Han, Kevin and Maddikayala, Siddharth and Knappe, Tim and Patel, Om and Liao, Austen and Barati Farimani, Amir , editor =. Proceedings of the 19th. 2026 , pages =. doi:10.18653/v1/2026.eacl-long.70 , abstract =

    work page doi:10.18653/v1/2026.eacl-long.70 2026

  16. [16]

    Attention is

    Vaswani, Ashish and Shazeer, Noam and Parmar, Niki and Uszkoreit, Jakob and Jones, Llion and Gomez, Aidan N and Kaiser, Lukasz and Polosukhin, Illia , year =. Attention is. Advances in

  17. [17]

    Artificial Intelligence Review , author =

    Human-in-the-loop machine learning: a state of the art , volume =. Artificial Intelligence Review , author =. 2023 , keywords =. doi:10.1007/s10462-022-10246-w , abstract =

    work page doi:10.1007/s10462-022-10246-w 2023

  18. [18]

    Additive Manufacturing , author =

    Accurate detection of local porosity in laser powder bed fusion through deep learning of physics-based in-situ infrared camera signatures , volume =. Additive Manufacturing , author =. 2025 , keywords =. doi:10.1016/j.addma.2025.104701 , abstract =

    work page doi:10.1016/j.addma.2025.104701 2025

  19. [19]

    Advances in Neural Information Processing Systems , author =

    Visual. Advances in Neural Information Processing Systems , author =. 2023 , pages =

    2023

  20. [20]

    The C Users Journal archive , author =

    A new algorithm for data compression , url =. The C Users Journal archive , author =

  21. [21]

    2025 , pages =

    Construction Robotics , author =. 2025 , pages =. doi:10.1007/s41693-025-00162-0 , abstract =

    work page doi:10.1007/s41693-025-00162-0 2025

  22. [22]

    Modular large language model agents for multi-task computational materials science , url =

    Chaudhari, Akshat and Ock, Janghoon and Barati Farimani, Amir , year =. Modular large language model agents for multi-task computational materials science , url =. Communications Materials , publisher =

  23. [23]

    Journal of Chemical Information and Modeling , author =

    Polymer-. Journal of Chemical Information and Modeling , author =. 2026 , pages =. doi:10.1021/acs.jcim.6c00343 , language =

    work page doi:10.1021/acs.jcim.6c00343 2026

  24. [24]

    Procedia Engineering , author =

    Selected. Procedia Engineering , author =. 2017 , keywords =. doi:10.1016/j.proeng.2017.02.245 , abstract =

    work page doi:10.1016/j.proeng.2017.02.245 2017

  25. [25]

    Materials & Design , author =

    Mechanical properties of components fabricated with open-source 3-. Materials & Design , author =. 2014 , keywords =. doi:10.1016/j.matdes.2014.02.038 , abstract =

    work page doi:10.1016/j.matdes.2014.02.038 2014

  26. [26]

    Additive Manufacturing , author =

    Tensile strength of commercial polymer materials for fused filament fabrication. Additive Manufacturing , author =. 2017 , keywords =. doi:10.1016/j.addma.2017.03.005 , abstract =

    work page doi:10.1016/j.addma.2017.03.005 2017

  27. [27]

    Pak, Peter , month = may, year =. ppak10/

  28. [28]

    Pak, Peter , month = apr, year =. ppak10/

  29. [29]

    2026 , note =

    slic3r/. 2026 , note =

    2026

  30. [30]

    2026 , note =

    openrocket/orhelper , copyright =. 2026 , note =

    2026

  31. [31]

    doi:10.1115/1.4067085 , abstract =

    Journal of Computing and Information Science in Engineering , author =. doi:10.1115/1.4067085 , abstract =

    work page doi:10.1115/1.4067085

  32. [32]

    Niskanen, Sampo , month = mar, year =

  33. [33]

    and Schimitt, Rodrigo N

    Ceotto, Giovani H. and Schimitt, Rodrigo N. and Alves, Guilherme F. and Pezente, Lucas A. and Carmo, Bruno S. , month = aug, year =. doi:10.1061/(ASCE)AS.1943-5525.0001331 , abstract =

    work page doi:10.1061/(asce)as.1943-5525.0001331 1943

  34. [34]

    and Nguyen, Andrew Q

    Sinclair, Bryce A. and Nguyen, Andrew Q. and Coker, Landon T. and Connor, Barrett J. and McLaughlin, Ida R. and Zander, Kevin , month = jul, year =. Low-. doi:10.2514/6.2025-3347 , language =

    work page doi:10.2514/6.2025-3347 2025

  35. [35]

    and Newman, James H

    Pierce, Dillon T. and Newman, James H. , month = mar, year =. Development and. Journal of Spacecraft and Rockets , publisher =. doi:10.2514/1.A36408 , abstract =

    work page doi:10.2514/1.a36408

  36. [36]

    and Fischer, U

    Kobald, M. and Fischer, U. and Tomilin, K. and Petrarolo, A. and Schmierer, C. , year =. Hybrid. Journal of Spacecraft and Rockets , publisher =. doi:10.2514/1.A34035 , abstract =

    work page doi:10.2514/1.a34035

  37. [37]

    2026 , note =

    anthropics/claude-code , url =. 2026 , note =

    2026

  38. [38]

    2026 , note =

    openai/codex , copyright =. 2026 , note =

    2026

  39. [39]

    2026 , note =

    google-gemini/gemini-cli , copyright =. 2026 , note =

    2026

  40. [40]

    Exploring the application of

    Zhou, Zhibin and Li, Yaoqi and Yu, Junnan , month = nov, year =. Exploring the application of. Human–Computer Interaction , publisher =. doi:10.1080/07370024.2024.2420991 , abstract =

    work page doi:10.1080/07370024.2024.2420991 2024

  41. [41]

    Ock, Janghoon and Sharma Meda, Radheesh and Vinchurkar, Tirtha and Jadhav, Yayati and Farimani, Amir Barati , year =. Adsorb-. Digital Discovery , publisher =. doi:10.1039/D5DD00298B , language =

    work page doi:10.1039/d5dd00298b

  42. [42]

    Catalyst

    Ock, Janghoon and Guntuboina, Chakradhar and Barati Farimani, Amir , month = dec, year =. Catalyst. ACS Catalysis , publisher =. doi:10.1021/acscatal.3c04956 , abstract =

    work page doi:10.1021/acscatal.3c04956

  43. [43]

    An LLM-based multi-agent system to assist early-stage product design and evaluation

    Jadhav, Yayati and Barati Farimani, Amir , month = feb, year =. Large language model agent as a mechanical designer , volume =. Journal of Engineering Design , publisher =. doi:10.1080/09544828.2026.2624356 , abstract =

    work page doi:10.1080/09544828.2026.2624356 2026

  44. [44]

    ACM Trans. Softw. Eng. Methodol. , author =. 2025 , pages =. doi:10.1145/3712003 , abstract =

    work page doi:10.1145/3712003 2025

  45. [45]

    2025 , keywords =

    IEEE Robotics and Automation Letters , author =. 2025 , keywords =. doi:10.1109/LRA.2025.3597835 , abstract =

    work page doi:10.1109/lra.2025.3597835 2025

  46. [46]

    doi:10.48550/arXiv.2504.19394 , abstract =

    Simonds, Toby , month = apr, year =. doi:10.48550/arXiv.2504.19394 , abstract =

    work page doi:10.48550/arxiv.2504.19394

  47. [47]

    2026 , note =

    prusa3d/. 2026 , note =

    2026

  48. [48]

    Markovsky, F

    Metal additive manufacturing in aerospace:. Materials & Design , author =. 2021 , keywords =. doi:10.1016/j.matdes.2021.110008 , abstract =

    work page doi:10.1016/j.matdes.2021.110008 2021

  49. [49]

    Advances in Neural Information Processing Systems , author =

    Toolformer:. Advances in Neural Information Processing Systems , author =. 2023 , pages =

    2023

  50. [50]

    2026 , note =

    Open-. 2026 , note =

    2026

  51. [51]

    doi:10.5281/ZENODO.3955118 , abstract =

    CadQuery contributors , month = feb, year =. doi:10.5281/ZENODO.3955118 , abstract =

    work page doi:10.5281/zenodo.3955118

  52. [52]

    and Lin, Kevin and Hewitt, John and Paranjape, Ashwin and Bevilacqua, Michele and Petroni, Fabio and Liang, Percy

    Lost in the. Transactions of the Association for Computational Linguistics , author =. 2024 , pages =. doi:10.1162/tacl_a_00638 , abstract =

    work page doi:10.1162/tacl_a_00638 2024

  53. [53]

    Cadsmith: Multi-agent cad generation with programmatic geometric validation.arXiv preprint arXiv:2603.26512, 2026

    Barkley, Jesse and Loghmani, Rumi and Farimani, Amir Barati , month = mar, year =. doi:10.48550/arXiv.2603.26512 , abstract =

    work page doi:10.48550/arxiv.2603.26512

  54. [54]

    Harry and Stine, Bill , year =

    Stine, G. Harry and Stine, Bill , year =. Handbook of model rocketry , isbn =

  55. [55]

    Winner, Robert and Pennell, James and Bertrand, Harold and Slusarczuk, Marko , year =. The

  56. [56]

    Kui, Wu and Chuan-jin, Gong , month = nov, year =. A. 2009. doi:10.1109/IITA.2009.187 , abstract =

    work page doi:10.1109/iita.2009.187 2009

  57. [57]

    Discover Mechanical Engineering , author =

    A systematic review of. Discover Mechanical Engineering , author =. 2025 , keywords =. doi:10.1007/s44245-025-00104-8 , abstract =

    work page doi:10.1007/s44245-025-00104-8 2025

  58. [58]

    Additive Manufacturing , author =

    Laser incidence angle influence on energy density variations, surface roughness, and porosity of additively manufactured parts , volume =. Additive Manufacturing , author =. 2022 , keywords =. doi:10.1016/j.addma.2021.102572 , abstract =

    work page doi:10.1016/j.addma.2021.102572 2022

  59. [59]

    Additive Manufacturing Letters , author =

    Laser shape variation influence on melt pool dynamics and solidification microstructure in laser powder bed fusion , volume =. Additive Manufacturing Letters , author =. 2023 , keywords =. doi:10.1016/j.addlet.2023.100141 , abstract =

    work page doi:10.1016/j.addlet.2023.100141 2023

  60. [60]

    , year =

    Bhat, Biliyar N. , year =. Aerospace

  61. [61]

    and Biblarz, Oscar and Morehart, James H

    Sutton, George P. and Biblarz, Oscar and Morehart, James H. , month = feb, year =. Rocket

  62. [62]

    , month = mar, year =

    Barrowman, J. , month = mar, year =. The

  63. [63]

    Acta Astronautica , author =

    Multidisciplinary design and optimization of winged architectures for reusable launch vehicles , volume =. Acta Astronautica , author =. 2023 , keywords =. doi:10.1016/j.actaastro.2023.05.041 , abstract =

    work page doi:10.1016/j.actaastro.2023.05.041 2023

  64. [64]

    Aerospace Science and Technology , author =

    Optimal air-launching rocket design using system trades and a multi-disciplinary optimization approach , volume =. Aerospace Science and Technology , author =. 2009 , keywords =. doi:10.1016/j.ast.2009.07.004 , abstract =

    work page doi:10.1016/j.ast.2009.07.004 2009

  65. [65]

    Aircraft Design , author =

    Launch vehicles conceptual design and structural analysis: an integrated approach via. Aircraft Design , author =. 1999 , pages =. doi:10.1016/S1369-8869(99)00011-7 , abstract =

    work page doi:10.1016/s1369-8869(99)00011-7 1999

  66. [66]

    2024 , keywords =

    FirePhysChem , author =. 2024 , keywords =. doi:10.1016/j.fpc.2023.10.001 , abstract =

    work page doi:10.1016/j.fpc.2023.10.001 2024

  67. [67]

    Acta Astronautica , author =

    Utilization of additive manufacturing in hybrid rocket technology:. Acta Astronautica , author =. 2021 , keywords =. doi:10.1016/j.actaastro.2020.11.024 , abstract =

    work page doi:10.1016/j.actaastro.2020.11.024 2021

  68. [68]

    2025 , keywords =

    Additive Manufacturing Letters , author =. 2025 , keywords =. doi:10.1016/j.addlet.2025.100292 , abstract =

    work page doi:10.1016/j.addlet.2025.100292 2025

  69. [69]

    LLM-3D print: Large Language Mod- els to monitor and control 3D printing

    Additive Manufacturing , author =. 2025 , keywords =. doi:10.1016/j.addma.2025.105027 , abstract =

    work page doi:10.1016/j.addma.2025.105027 2025

  70. [70]

    Additive Manufacturing Letters , author =

    Agentic additive manufacturing alloy evaluation , volume =. Additive Manufacturing Letters , author =. 2026 , keywords =. doi:10.1016/j.addlet.2026.100355 , abstract =

    work page doi:10.1016/j.addlet.2026.100355 2026