I am a Staff Applied Scientist at NetApp. Over 17 years, my research and engineering work has spanned deep learning, multi-agent AI systems, data science, and high-performance enterprise systems. I specialize in designing and deploying scalable intelligent systems, anomaly detection, and hardware-accelerated inference—bridging deep mathematical theory with rigorous systems engineering to ship production-grade AI under strict latency and compute constraints.
Currently, I lead an AI & ML R&D team at NetApp. My role is to bring frontier AI research into real enterprise systems. Before NetApp, I was a Data Scientist at DeepCell, developing deep learning models end-to-end through production deployment. I also served as a Researcher in Daniel Acuña’s Science of Science & Computational Discovery Lab (Syracuse University iSchool, 2020–2021), where we co-authored a Nature Portfolio paper on the lifespan of scientific resources, combining rigorous statistical modeling with large-scale metadata mining.
This blend of mathematical intuition and low-level systems engineering background (C++ and Java) lets me span the critical stages of the ML lifecycle—from custom algorithms to high-throughput inference deployment on GPUs and TPUs with PyTorch, TensorFlow, TensorRT, and ONNX. For a full account of my 17-year path from traditional systems development to applied AI research, see Career Journey. In navigating that transition, reaching one’s destination is like a long journey, yet the road itself feels like driving on a highway at night—you can only see the stretch immediately ahead. So I have learned to live in the present: instead of worrying about what I cannot yet see, I focus on the next stretch of road; the way forward clarifies itself as you keep going.
Research, Patents & Open-Source
The following work spans statistical modeling and reproducible research through to production-grade ML systems and patent-backed deployment:
1. Publication & Open Source
- Predicting the longevity of resources shared in scientific publications (Published in Humanities and Social Sciences Communications, Nature Portfolio, 2025). DOI: 10.1057/s41599-025-04716-z
- Authors: Daniel E. Acuña, Jian Jian, Tong Zeng, Lizhen Liang, Han Zhuang.
- Funding: Supported by the U.S. Office of Research Integrity (Project Grants: ORIIR190049 and ORIIR180041).
- Core open-source tool: I independently developed and open-sourced Tobit-EN, a Python package implementing Tobit censored regression with Elastic Net regularization via maximum a posteriori (MAP) estimation to model and forecast the lifespan of digital resources in scientific literature.
2. Patents (5 Published Patents & 2 Applications)
- As a key member of the core R&D team, I co-developed and co-invented multiple patents that form the algorithmic foundation of NetApp ONTAP’s real-time Autonomous Ransomware Protection (ARP) engine (featured in Forbes):
- Graph Vector Variation Driven Data Corruption Detection (US Patent Pub. No: US20250245326A1)
- Vector Variation Driven Malware Corruption Detection (US Patent Pub. No: US20250245324A1)
- Malicious encryption detection based on byte frequency distribution (US Patent Pub. No: US20250298892A1)
- Data Exfiltration Monitoring Using Hash Values (US Patent Pub. No: US20250330488A1)
- Data Exfiltration Monitoring Using Semantic Queries (US Patent Pub. No: US20250330487A1)
- Protection Graph (Pending Application)
- Clean Room Mechanism (Pending Application)
3. Open Source Research Tools
- ImageAnnotatorJS: A modular JavaScript library (AMD standard) for building academic figure inspection and annotation systems, funded by the U.S. Office of Research Integrity (ORI).
Beyond the Code
World of Warcraft
Badminton
Community Fundraising
Click to view donor roster
