# YIPENG WANG - AI Research & Engineering Leader | Multimodal, World Models
Source: https://hello.cv/yipengwang
yipeng.wang.99@outlook.com | (314) 680-2939 | Seattle

## Links
- LinkedIn | https://www.linkedin.com/in/yipeng-wang99
- X | https://x.com/chill_chio
- GitHub | https://github.com/patina27
- Email | yipeng@pelagornis.ai

## About
Highly accomplished AI Research & Engineering Leader with a proven track record in developing cutting-edge generative models, 3D vision systems, and large-scale data pipelines. Expert in driving state-of-the-art advancements in image and video generation, 3D reconstruction, and neural fields, with significant contributions to leading industry platforms. Adept at leading research teams and delivering high-impact technical solutions from conception to deployment, poised to excel in advanced AI/ML research and engineering roles.

## Work
### Member of Technical Staff | Microsoft Superintelligence Team
Leading the development of agentic capabilities for MAI-Image image generation models, achieving top rankings in competitive leaderboards.
- Led development of agentic capabilities for MAI-Image imagegen models, securing a top 3 position in the LMArena Leaderboard, trailing only OpenAI's GPT-Image-2 and Deepmind's NanoBanana 2.
- Contributed to advancing the state-of-the-art in agentic image generation, demonstrating expertise in complex AI system design and optimization.
### Member of Technical Staff | World Models, World Labs
Spearheaded research and development for 3D scene generation, encompassing data strategy, model training, and advanced data curation pipelines.
- key contributor to the model training of Marble 1 - the world's first foundation model for 3D scene generation, powered by multi-view generative model.
- Directed data strategy and led real-world spatial data collection efforts and PB-level internet multimodal data sourcing, establishing a comprehensive data pyramid strategy for pre-training and post-training spatial 3D datasets.
- Investigated reinforcement learning methods to improve view-consistency of the multi-view diffusion model, enhancing model performance and realism.
### Research Engineer | Pika Labs
Contributed significantly to the development of Pika 1.5 and 2.0, focusing on data curation, distributed training, and advanced video generation features.
- Served as a main contributor to Pika 1.5 and 2.0, with the latter scoring #5 in the artificial analysis leaderboard.
- Developed a ray-based heterogeneous data curation pipeline that processed 5PB of video data, yielding billions of video clips for training.
- Engineered a large-scale distributed training framework utilizing FSDP+CP, optimizing computational efficiency and resource utilization.
- Led post-training research on camera-pose conditioned video generation and ingredients-to-video features, enhancing model control and output quality.
### CV Engineer II | Meta Reality Labs
Developed and optimized a pipeline for generating detailed 3D models from user-captured media for real-time rendering on standalone VR headsets.
- Created a pipeline for generating detailed 3D models from user-captured images or videos, optimized for real-time rendering on standalone VR headsets.
- Utilized NeRF-based reconstruction and advanced baking techniques to achieve high-fidelity and efficient 3D model generation.
- Directed pioneering research initiatives to enable object interaction within user-captured scenes, expanding immersive VR experiences.
### CV Engineer | Meta Reality Labs
Engineered innovative algorithms to transform 2D monocular images and videos into 3D formats, enabling the creation of immersive 3D content.
- Engineered innovative algorithms to transform 2D monocular images and videos into 3D formats, enabling the creation of 3D photos, 3D videos, and stereo videos.
- Utilized NeRF-based novel-view synthesis and mesh baking & streaming techniques to enhance visual quality and delivery efficiency.

## Education
### Washington University in St. Louis | Computer Science
4.00
### Washington University in St. Louis | Computer Science + Mathematics
3.97

## Awards
### Ranked No. 38, ACM-ICPC North America Championship
ACM-ICPC | 2020-01-01
Achieved a top ranking in the prestigious North America Championship for competitive programming.
### Ranked No. 6, ACM-ICPC Mid-Central USA Regional Contest
ACM-ICPC | 2018-01-01
Secured a high ranking in the Mid-Central USA Regional Contest for competitive programming.
### Bronze Award, Asia-Pacific Informatics Olympiad (China Division)
Asia-Pacific Informatics Olympiad | 2016-01-01
Received a Bronze Award in the highly competitive Asia-Pacific Informatics Olympiad, representing China.
### First Prize, National Olympiad in Informatics in Provinces (NOIP), Senior Group
National Olympiad in Informatics in Provinces | 2015-01-01
Awarded First Prize in the Senior Group of the National Olympiad in Informatics in Provinces.

## Publications
### Reflection-Aware Neural Radiance Fields
SIGGRAPH ASIA
Co-authored a publication on Reflection-Aware Neural Radiance Fields, contributing to advancements in realistic rendering and scene reconstruction.
### LTM: Lightweight Textured Mesh Extraction and Refinement of Large Unbounded Scenes for Efficient Storage and Real-time Rendering
CVPR
Contributed to research on Lightweight Textured Mesh Extraction and Refinement (LTM) for efficient storage and real-time rendering of large scenes.
### OmnimatteRF: Robust Omnimatte with 3D Background Modeling
ICCV
Co-authored research on OmnimatteRF, focusing on robust omnimatte generation with advanced 3D background modeling techniques.

## Source
Read this profile on Hello.cv: https://hello.cv/yipengwang
Create your free profile at https://hello.cv