Undergraduate at Nanjing University (VLSI Design & System Integration), graduating in 2027.
I work on autonomous AI systems, AI accelerators, computer architecture, and hardware-software co-design.
Personal website: ahappend.github.io
- Nanjing University — VLSI Design & System Integration
- Campus locations: Nanjing (Year 1); Suzhou (Years 2–4)
- Expected graduation: 2027
| Period | Experience |
|---|---|
| Sep–Dec 2026 | Microsoft Research Asia (MSRA) — Research Intern, Systems Research Group, Beijing. Group website ↗ |
| Jan–May 2027 | Nanyang Technological University (NTU), Singapore — Government-Sponsored Visiting Undergraduate. Program website ↗ |
Autonomous AI DNN Mapping Scheduling NoC RTL Edge AI Reproducible Systems
- Native AWQ W4A16 and mixed-precision RTL for transformer inference
- DNN mapping and inter-layer scheduling for tiled accelerators
- Hardware-software co-design for AI accelerators
- Evidence-driven autonomous research and engineering systems
- Argus/original / ArgusAgent/Microsoft — Open multi-agent runtime for long-horizon research and engineering, combining persistent state, reviewed execution, verification, and evidence. Technical report ↗
- ACE-3/personal / ACE-3/organization — Argus-built, evidence-first Qwen2.5-0.5B-Instruct-AWQ W4A16 RTL with a verified 24-layer cascade and authenticated Hybrid RTL runtime.
- ACE-2/personal / ACE-2/organization — Argus-built Qwen2.5-0.5B W4A8 accelerator with SystemVerilog RTL, certified two-token integration, SKY130 100 MHz results, and evidence-first host-trust recovery.
- 64-Point FFT — Resource-efficient radix-2 DIT FFT in SystemVerilog using folded streaming SDF RTL, Q1.15 fixed-point arithmetic, Vivado/XSim simulation, Python/NumPy co-verification, and Virtex-7 implementation results.
- Two-Stage CMOS Op-Amp — Two-stage CMOS operational amplifier in TSMC 0.18 μm technology, with Miller compensation and Spectre AC/DC/transient verification.
- HyperCut: Fast Inter-Layer Scheduling via Directed Hypergraph and Early Filtering — Early filtering and hypergraph-based inter-layer scheduling for tiled DNN accelerators.
- Argus: A General-Purpose Agentic Reasoning Runtime for Long-Horizon Tasks — A persistent, self-evolving runtime for long-horizon agentic reasoning.
Python C/C++ TypeScript SystemVerilog Verilog React Linux