I build database engines and agent infrastructure from the ground up. As the primary creator and core contributor behind four major open-source systems, I focus on solving the hardest challenges in vector search, hyper-convergence, memory-centric storage, and durable agent execution.
-
🔭 Currently Building (Core Kernel):
- MatrixOne: Driving the core architecture and kernel evolution of the hyper-converged cloud-native database.
- Memoria: Leading the foundational design and implementation of the next-gen memory-centric storage engine.
- Astra: Building the context-to-execution runtime for durable, governed enterprise agents.
-
🌟 Created From Scratch:
- Milvus: Founded and built the world's most popular open-source vector database, powering AI applications for thousands of enterprises.
-
🧠 Deep Expertise:
Database Kernel InternalsDistributed ConsensusVector Search AlgorithmsStorage Engine DesignAgent Runtime DesignContext EngineeringGoRustC++ -
💬 Ask me about: Building a DBMS from zero, scaling vector search to billions, HTAP architecture, durable agent runtimes, or the internals of Milvus/MatrixOne/Memoria/Astra.
| Project | Description |
|---|---|
| MatrixOne | Hyper-converged cloud-native database merging OLTP, OLAP, and Stream processing. |
| Memoria | Next-generation memory-centric storage engine for modern workloads. |
| Astra | Context-to-execution runtime for durable, governed enterprise agents. |
| Milvus | The original cloud-native vector database for AI, created from scratch. |





