AIML + Product + Software Development + Research
All-rounded engineer + scientist. I can literally do anything for early AI startups: Agentic Systems, ML/DL/RL model training, full stack + infra, AI research
- Physical AI: PhD in Systems Engineering (Cyber-physical Systems) from UC Berkeley, with foundational experience in robotics and self-driving car projects (2017-2018) including a published paper on attention maps at ACCV. Since steering back to physical AI and robotics in early 2025, now bridging my expertise in spatial systems, transformer-based model training, and agentic systems to work on embodied agents for robotic task planning and VLA for robotic manipulation.
- DL/RL model training: I was among the first batch to get hands-on experience with transformers (2018), Hugging Face transformers (2019), and LLMs (2021).
- Agentic Systems: (LiteMultiAgent/ LiteWebAgent/ VisualTreeSearch open source repos, ~200 stars)
- Full-stack + Infrastructure: Independently developing and deploying complete web applications (2025-present) with Next.js frontends on Vercel and Express.js/FastAPI backends on AWS ECS
- AI research: 20+ AI/ML papers, 700+ citations, founded a non-profit AI research community and wrote 2 papers (one accepted to NAACL 2025, one accepted to ECML-PKDD 2025)
