Back to dashboard

Spectral is building generative foundation models that turn user input into stru…

completed38 qualified1 runMay 5, 7:20 AMspectral-is-building-generative-foundation-models-that-turn-1777965657
ParsedSpectral · 3 topics · Senior · Hybrid
Generating seed nodes
0 proposed
Explored 0 queries
0/0 done
    3
    Expanding nodes
    queued
    4
    Qualifying candidates
    queued

    Qualified Candidates (32)

    FZ

    Fuyang Zhang

    high hireability

    Research Assistant@Simon Fraser University

    Previously: Intern @ Meta

    Burnaby, CA

    79
    3D Geometry Representations85
    Geometric Deep Learning78
    Generative CAD Models75
    Strengths
    BR-DF (2025): B-rep via volumetric distance functions — direct CAD relevance
    SceneScript (ECCV 2024): autoregressive structured scene model, analogous to CAD generation
    Gaps
    No direct BrepGen/SkexGen co-authorship — adjacent lab, not the core CAD papers
    …click to see all
    JC

    Jiacheng Chen

    high hireability

    PhD Candidate@Simon Fraser University

    Previously: Ph.D. student @ University of Southern California

    Vancouver, CA

    57
    Geometric Deep Learning65
    3D Geometry Representations55
    Generative CAD Models52
    Strengths
    Floor-SP: inverse CAD via sequential shortest path (ICCV 2019, 126 cites)
    PolyDiffuse: guided set diffusion for structured shape reconstruction (NeurIPS 2023)
    Gaps
    Structured outputs are 2D (floorplans, HD maps) — no B-rep, feature tree, or NURBS work
    …click to see all
    MT

    Maham Tanveer

    high hireability

    Ph.D. student@Simon Fraser University

    Previously: Software Engineer @ Sedenius Technologies

    CA

    68
    3D Geometry Representations72
    Generative CAD Models70
    Geometric Deep Learning62
    Strengths
    D²CSG NeurIPS 2023 (43 citations): neural CSG tree learning for 3D CAD
    CSG trees — directly a structured/editable CAD representation
    Gaps
    Recent work pivoted to video inbetweening — away from 3D CAD
    …click to see all
    SX

    Sam Xu

    high hireability
    98
    Generative CAD Models100
    3D Geometry Representations98
    Geometric Deep Learning95
    Strengths
    BrepGen (SIGGRAPH 2024): diffusion-based B-rep generative model
    AutoBrep (SIGGRAPH Asia 2025): autoregressive Transformer for unified B-rep tokens
    Gaps
    Canada-based; Spectral JD has no location constraint but worth noting
    …click to see all
    SX

    Sam Xu

    high hireability
    97
    Generative CAD Models100
    3D Geometry Representations98
    Geometric Deep Learning92
    Strengths
    BrepGen (SIGGRAPH 2024): diffusion over hierarchical B-rep latent tree
    AutoBrep (SIGGRAPH Asia 2025): autoregressive Transformer for B-rep tokens
    Gaps
    Work is primarily diffusion/autoregressive — limited public RL-for-generative-models experience
    …click to see all
    SX

    Sam Xu

    high hireability
    97
    Generative CAD Models99
    3D Geometry Representations97
    Geometric Deep Learning95
    Strengths
    BrepGen SIGGRAPH 2024: diffusion model over hierarchical B-rep latent tree
    AutoBrep SIGGRAPH Asia 2025: autoregressive Transformer BFS-ordered B-rep tokens on ABC-1M
    Gaps
    Based in Canada — no US presence unless relocating
    …click to see all
    AS

    Aditya Sanghi

    medium hireability

    Director@Samruddhi

    Previously: Director @ Sanghi Industries Ltd

    Ahmedabad, IN

    93
    Generative CAD Models97
    3D Geometry Representations93
    Geometric Deep Learning88
    Strengths
    SolidGen: autoregressive B-rep synthesis — core match to Spectral's SGS work
    BRepNet: topological message passing on solid model B-rep faces/edges
    Gaps
    No explicit open-to-work signals — pipeline shows no LinkedIn/website changes
    …click to see all
    AB

    Aljaz Bozic

    medium hireability

    Research Scientist@Meta

    Previously: Research Scientist @ Meta

    Zurich, CH

    58
    Geometric Deep Learning82
    3D Geometry Representations78
    Generative CAD Models15
    Strengths
    NPMs: Neural Parametric Models — parametric 3D shape space learning
    Neural Deformation Graphs (CVPR 2021 Oral) — graph-based 3D deep learning
    Gaps
    No CAD-specific work — no B-rep, feature trees, or parametric structure generation
    …click to see all
    HG

    Hao-Xiang Guo

    medium hireability
    86
    3D Geometry Representations90
    Geometric Deep Learning85
    Generative CAD Models82
    Strengths
    ComplexGen (SIGGRAPH 2022): B-rep reconstruction from point clouds, 125 GitHub stars
    NH-Rep (SIGGRAPH Asia 2022): neural implicit conversion of B-rep solids
    Gaps
    No parametric/feature-tree CAD or editable B-rep generation — reconstruction focus
    …click to see all
    KH

    Ka-Hei Hui

    medium hireability

    Researcher@Autodesk

    Previously: PhD student @ The Chinese University of Hong Kong

    US

    55
    3D Geometry Representations75
    Geometric Deep Learning65
    Generative CAD Models25
    Strengths
    SP-GAN (164 cites) — landmark 3D shape generation paper
    Neural Wavelet Diffusion (144 cites) — 3D diffusion generative model
    Gaps
    No published B-rep, feature tree, or parametric CAD papers
    …click to see all
    KM

    Kamal Rahimi Malekshan

    medium hireability

    Principal Machine Learning Engineer@Autodesk

    Previously: Senior Research Engineer - Machine Learning @ Autodesk

    Toronto, CA

    79
    Generative CAD Models88
    3D Geometry Representations78
    Geometric Deep Learning72
    Strengths
    Zero-to-CAD (Apr 2025): agentic parametric CAD program synthesis at 1M scale
    Editable Prismatic CAD (2022): structured, editable CAD output from voxels
    Gaps
    No evidence of B-rep or NURBS-specific ML work
    …click to see all
    KM

    Kanika Madan

    medium hireability

    Principal Researcher@Autodesk

    Previously: Senior Manager Business Solutions @ AU

    Gurugram, IN

    57
    3D Geometry Representations65
    Generative CAD Models55
    Geometric Deep Learning50
    Strengths
    WaLa: billion-parameter 3D generative model, wavelet latent diffusion (2024)
    Image-to-3D texture mapping with triplane representation (2025)
    Gaps
    3D work is SDF/texture-based, not structured CAD (B-rep, feature trees, parametric)
    …click to see all
    MF

    Marco Fumero

    medium hireability

    Postdoctoral Researcher@Institute of Science and Technology Austria

    Previously: Doctoral Researcher @ Sapienza Università di Roma

    Vienna, AT

    49
    Geometric Deep Learning75
    3D Geometry Representations55
    Generative CAD Models18
    Strengths
    GLADIA lab PhD under Rodolà — top spectral GDL group globally
    CLIP-Forge co-author — generative 3D shape from text (zero-shot)
    Gaps
    No B-rep, NURBS, or parametric feature-tree experience
    …click to see all
    NM

    Nigel J. W. Morris

    medium hireability

    Autodesk

    Previously: Researcher @ Autodesk

    71
    Generative CAD Models82
    3D Geometry Representations72
    Geometric Deep Learning58
    Strengths
    SolidGen (TMLR 2023): autoregressive direct B-rep synthesis — on-point
    Indexed B-rep representation: vertices/edges/faces hierarchy for ML
    Gaps
    No LinkedIn/GitHub — limited discoverability and outreach path
    …click to see all
    ND

    Nishkrit Desai

    medium hireability

    Researcher@Axiom

    Previously: Intern @ NVIDIA

    87
    Generative CAD Models95
    3D Geometry Representations90
    Geometric Deep Learning75
    Strengths
    SolidGen: autoregressive model directly synthesizing B-rep CAD (ICLR 2024)
    Invented Indexed B-rep for ML — topological hierarchy for transformers
    Gaps
    No LinkedIn — tenure at Axiom unknown, hireability signals limited
    …click to see all
    PJ

    Pradeep Kumar Jayaraman

    medium hireability
    94
    Generative CAD Models98
    3D Geometry Representations95
    Geometric Deep Learning90
    Strengths
    BrepGen (SIGGRAPH 2024) — diffusion generative model for B-rep CAD
    AutoBrep (SIGGRAPH Asia 2025) — autoregressive B-rep with topology
    Gaps
    Long Autodesk tenure (~7 yrs) — no visible mobility or open-to-work signals
    …click to see all
    RG

    Ruiqi Gao

    medium hireability

    Staff Research Scientist@DeepMind

    Previously: Research Scientist @ Google

    San Francisco, US

    27
    3D Geometry Representations40
    Geometric Deep Learning35
    Generative CAD Models5
    Strengths
    CAT3D (2024): multi-view diffusion model for 3D generation at scale
    Bolt3D (2025): 3D scene generation with latent diffusion, 300x faster
    Gaps
    No CAD work — no B-rep, NURBS, feature trees, or STEP file experience
    …click to see all
    XY

    Xingguang Yan

    medium hireability

    PhD student@Simon Fraser University

    Previously: Research Intern @ NVIDIA

    CA

    50
    3D Geometry Representations68
    Geometric Deep Learning65
    Generative CAD Models18
    Strengths
    Omages: Best Paper 3DV 2025 — novel 3D generation via structured mesh representation
    ShapeFormer: Transformer on sparse 3D point clouds for shape completion
    Gaps
    No CAD-specific work — no B-rep, feature trees, STEP, or parametric structure
    …click to see all
    YL

    Yang Liu

    medium hireability

    Principal Researcher@Microsoft

    Previously: Postdoctoral Fellow @ LORIA/INRIA

    China

    86
    3D Geometry Representations92
    Geometric Deep Learning85
    Generative CAD Models82
    Strengths
    ComplexGen (SIGGRAPH 2022): B-rep chain complex generation for CAD reconstruction
    Neural Halfspace Representation for manifold B-rep solids (2022)
    Gaps
    15+ years at MSRA — long tenure creates recruitment friction
    …click to see all
    ZW

    Zhengqing Wang

    medium hireability

    Ph.D. student@Simon Fraser University

    Montreal, CA

    77
    Generative CAD Models82
    3D Geometry Representations78
    Geometric Deep Learning72
    Strengths
    BrepGen SIGGRAPH 2024 — B-rep diffusion model, 76 citations, directly on-query
    PuzzleFusion++ ICLR 2025 — 3D fracture assembly via denoising/verification
    Gaps
    Single CAD-specific paper (BrepGen); recent work shifted to autonomy/video
    …click to see all
    AB

    Adrien Bousseau

    low hireability

    Researcher@INRIA

    Previously: Postdoc @ UC Berkeley

    Nice, FR

    76
    Generative CAD Models85
    Geometric Deep Learning72
    3D Geometry Representations70
    Strengths
    Free2CAD (2022) — generates structured CAD command sequences from drawings
    Sketch2CAD (2020) — sequential structured CAD modeling via deep learning
    Gaps
    No large-scale generative model work (diffusion, autoregressive, LLM-scale)
    …click to see all
    AJ

    Alec Jacobson

    low hireability

    Researcher@Adobe

    Previously: Senior Research Scientist @ Adobe

    Toronto, CA

    69
    3D Geometry Representations93
    Geometric Deep Learning83
    Generative CAD Models30
    Strengths
    libigl creator — foundational C++ geometry processing library used industry-wide
    CSG on neural signed distance fields (2023) — directly relevant to CAD solid modeling
    Gaps
    No direct generative CAD (B-rep or feature-tree generation) published work
    …click to see all
    AR

    Arianna Rampini

    low hireability

    Senior Research Scientist@Autodesk

    Previously: Junior Research Scientist @ Autodesk

    IT

    63
    Geometric Deep Learning85
    3D Geometry Representations75
    Generative CAD Models30
    Strengths
    Make-A-Shape: 10M-scale 3D diffusion (wavelet-tree SDF), ICML 2024
    WaLa: billion-param 3D generative with 2427x SDF compression
    Gaps
    No published work on structured parametric CAD (B-rep, feature trees, STEP)
    …click to see all
    CQ

    Charles R. Qi

    low hireability

    Member of Technical Staff@OpenAI

    Previously: Staff Machine Learning Scientist @ Tesla

    San Francisco, US

    62
    Geometric Deep Learning92
    3D Geometry Representations82
    Generative CAD Models12
    Strengths
    PointNet (21K citations) — created foundational 3D point cloud DL architecture
    PointNet++ (16K citations) — hierarchical feature learning on point sets
    Gaps
    No generative CAD/B-rep/NURBS/parametric structure experience
    …click to see all
    HP

    Hao Pan

    low hireability

    Assistant Professor@Tsinghua University

    Previously: Researcher @ Microsoft

    Beijing, CN

    89
    Generative CAD Models95
    3D Geometry Representations90
    Geometric Deep Learning83
    Strengths
    ComplexGen: B-rep chain complex generation (SIGGRAPH 2022, 126 citations)
    Sketch2CAD + Free2CAD: sequential/autoregressive CAD modeling (SIGGRAPH 2021/2022)
    Gaps
    New Tsinghua AP (Sep 2024) — unlikely to leave academia near-term
    …click to see all
    HS

    Hooman Shayani

    low hireability

    Principal Researcher@Autodesk

    London, GB

    74
    3D Geometry Representations88
    Geometric Deep Learning82
    Generative CAD Models52
    Strengths
    BRepNet (CVPR 2021, 159 cit) — topology message passing on B-rep solids
    UV-Net (CVPR 2021, 125 cit) — deep learning on B-rep boundary representations
    Gaps
    B-rep papers focus on learning/segmentation, not structured generative CAD with feature trees
    …click to see all
    JL

    Joe Lambourne

    low hireability
    94
    Generative CAD Models98
    3D Geometry Representations95
    Geometric Deep Learning90
    Strengths
    BrepGen (SIGGRAPH 2024) — diffusion model for B-rep CAD generation
    AutoBrep (SIGGRAPH Asia 2025) — autoregressive B-rep with topology+geometry tokens
    Gaps
    Very long tenure at Autodesk (15+ years) — strong entrenched signal
    …click to see all
    MF

    Matthew Fisher

    low hireability

    Principal Scientist@Adobe

    Previously: Graduate student @ Stanford University

    61
    3D Geometry Representations82
    Geometric Deep Learning80
    Generative CAD Models20
    Strengths
    ShapeShifter (CVPR 2025): 3D shape variation via multiscale point-voxel diffusion
    RenderDiffusion (CVPR 2023): diffusion for 3D reconstruction — 293 citations
    Gaps
    No explicit CAD/B-rep/parametric feature-tree work in publication record
    …click to see all
    PG

    Paul Guerrero

    low hireability

    Research Scientist@Adobe

    Previously: Research Associate (Post-Doc) @ University College London

    London, GB

    83
    Geometric Deep Learning90
    3D Geometry Representations85
    Generative CAD Models75
    Strengths
    SketchGen (NeurIPS 2021) — generates constrained CAD sketches directly
    StructureNet — hierarchical graph nets for 3D shape generation (joint 1st author)
    Gaps
    No work on B-rep, feature trees, or solid CAD — 3D generative is sketch-level
    …click to see all
    QC

    Qimin Chen

    low hireability

    PhD student@Simon Fraser University

    Previously: Research Scientist/Engineer Intern @ Adobe

    Burnaby, CA

    69
    3D Geometry Representations80
    Geometric Deep Learning72
    Generative CAD Models55
    Strengths
    D²CSG (NeurIPS 2023): learns compact CSG trees for 3D CAD shapes
    GenVDM (CVPR 2025 highlight): generates parametric VDMs from single image
    Gaps
    No direct work on B-rep, feature trees, or parametric CAD workflows
    …click to see all
    TG

    Thibault Groueix

    low hireability

    Research Scientist@Meta

    Previously: Research Scientist @ Adobe

    San Francisco, US

    74
    Geometric Deep Learning92
    3D Geometry Representations88
    Generative CAD Models42
    Strengths
    AtlasNet CVPR 2018 — 1678 citations, seminal 3D surface generation
    Instant3dit CVPR 2025 — multiview inpainting for 3D editing
    Gaps
    No visible work with parametric/structured CAD (B-rep, feature trees, STEP)
    …click to see all
    XT

    Xin Tong

    low hireability

    Partner Research Manager@Anuttacon

    Previously: Partner Research Manager @ Microsoft

    San Francisco, US

    86
    3D Geometry Representations95
    Geometric Deep Learning90
    Generative CAD Models72
    Strengths
    ComplexGen (SIGGRAPH 2022) — B-rep chain complex generation for CAD
    TRELLIS (2025, 297 citations) — structured 3D latent generative model
    Gaps
    ComplexGen is reconstruction, not forward generative CAD synthesis
    …click to see all

    Runs

    #1completed0 qualified / 0 foundMay 5, 7:20 AM