[NODE: INIT]
PID: 000 | STATUS: RUNNING
// Munich, Germany
[NODE: DATA_INPUT]
PID: 010 | STATUS: STREAMING
Founding Engineer at Kyrall, building AI-powered 3D CAD generation systems. PhD researcher in Generative AI for Engineering Design, specializing in deep generative models that bridge the gap between machine learning and real-world product design. My work spans spectral-domain diffusion models, NURBS surface learning, and plausibility evaluation of AI-generated designs. Published at ICCV, ECCV, IEEE CAI, and arXiv. Passionate about turning cutting-edge research into production systems.
[NODE: PROCESS] — CENTRAL
PID: 100 | STATUS: COMPLETED
Deep Generative Models for Engineering Design
Successfully defended on 2026-03-24 at Leiden University. Dissertation focused on bridging foundational AI research and industrial practice — developing generative models that produce structurally meaningful, CAD-ready geometry usable in real engineering workflows, with applications in automotive design at BMW.
[NODE: OUTPUT[]]
PID: 200 | COUNT: 6
-
SpoDify"A Shape Is Worth 512 Numbers: Spectral-domain Diffusion Modeling for 3D Shape Generation"IEEE CAI 2026 -
FDD"Enhancing Plausibility Evaluation for Generated Designs with Denoising Autoencoder"ECCV 2024 -
NeuroNURBS"NeuroNURBS: Learning Efficient Surface Representations for 3D Solids"Under Review -
PODM"On the Noise Scheduling for Generating Plausible Designs with Diffusion Models"arXiv 2023 -
SA-ALAE"Adversarial latent autoencoder with self-attention for structural image synthesis"IEEE CAI 2024 -
GeoDiffusion"A Training-Free Framework for Accurate 3D Geometric Conditioning in Image Generation"ICCV 2025
[NODE: RUNTIME_LOG]
PID: 300 | ENTRIES: 5
-
Founding Engineer Aug 2025 – Present | MunichKyrallBuilt AI-powered 3D CAD generation system; agentic workflows; instruction management; data creation infrastructure.
-
PhD Researcher Oct 2022 – Jul 2025 | MunichBMW GroupPublished 6 papers (ECCV 2024, IEEE CAI 2024, IEEE CAI 2026); developed NeuroNURBS; BMW Doctoral Speaker 2024.
-
Internship Apr – Nov 2021 | MunichBMW GroupAR/VR development; digital twin projects.
-
Full-stack Developer May 2021 – Oct 2022 | MünsterBausichtTypeScript/Vue.js web application development.
-
Working Student Apr 2020 – Feb 2021TU DarmstadtRNN/LSTM time-series forecasting research.
[NODE: CONFIG]
PID: 400 | MODULES: LOADED
LLM / AI Engineering
Generative ML
CAD / 3D Modeling
Data Engineering
Cloud / Infra
Python
C#
TypeScript
Systems Architecture
[NODE: MODULES]
PID: 500 | STATUS: LOADED
-
PhD, Advanced Computer ScienceLeiden UniversityOct 2022 – Aug 2025
-
MSc, Mechanical EngineeringTU DarmstadtOct 2018 – Oct 2022
[NODE: EXIT]
PID: 999 | AWAITING_INPUT
// Establish connection
EMAIL: jiajie.jf@gmail.com
GITHUB: jiajie96 (PhD)
GITHUB: JiajieAI (Kyrall)
LINKEDIN: /in/jiajiefan
GOOGLE SCHOLAR
> PROCESS TERMINATED SUCCESSFULLY. EXIT CODE 0