AIML + Robotics + Product + Software Development + Research
All-rounded engineer + scientist. I build everything end-to-end for early AI startups — from agentic systems, ML/DL/RL model training, and AI research, to robotics, full-stack infra, and edge deployment.
- Physical AI & Robotics: PhD in Systems Engineering (Cyber-Physical Systems) from UC Berkeley. Experienced in robotics and self-driving systems (2017–2018, ACCV paper on attention maps). Since 2025, re-focusing on embodied agents and robotic task planning — bridging transformers, spatial reasoning, and RL-based manipulation.
- Platforms:
ROS2,WebRTC,Socket.IO,LeRobot - Edge & Devices:
Raspberry Pi,myCobot 280,Hiwonder JetAuto,SO-101 - 3D Design & Fabrication:
Blender,Bambu Lab(3D printing workflows)
- Platforms:
- DL/RL Model Training: Early adopter of Transformers (2018), Hugging Face (2019), and LLMs (2021).
- Agentic Systems: Creator of LiteMultiAgent, LiteWebAgent, VisualTreeSearch (200+ stars total)
- Full-stack + Infrastructure: Next.js frontends (Vercel) + Express/FastAPI backends (AWS ECS)
- AI Research: 20+ papers, 700+ citations, founded a non-profit AI research community (papers accepted to NAACL 2025 and ECML-PKDD 2025)





