You can call me Boy, as it sounds similar to my Chinese name.
M.S. Computer Science candidate at National Cheng Kung University, focused on systems and network software, concurrency, and performance engineering. Open-source contributor to Hummingbot.
- 🎓 Expected graduation: August 2027 · Available to start: September 2027
- 👤 Website: matthiola.dev
- 📫 How to reach me: matthiola020@gmail.com
- M.S. in Computer Science — National Cheng Kung University, Tainan, Taiwan
- Expected Aug 2027
- B.S. in Interdisciplinary Program of Science — National Tsing Hua University, Hsinchu, Taiwan
- 2025
Hummingbot — upstream contributions
- ✅ Merged — #8338: Ensure candle requests always span at least one interval
Fixed zero-width historical candle requests caused by
limit=0on sub-interval ranges and pagination boundaries. Clamped each request to at least one candle interval, added boundary-candle deduplication, and introduced six regression tests. The complete candles test suite passed with 1,263 tests.
🔍 Pull requests currently under review
-
Under review — #8410: Prevent an XEMM maker-order race between the control loop and fill events Proposed a lifecycle-safe cancellation flow that keeps the tracked maker order until exchange confirmation, so stale asynchronous work cannot cancel an already-filled order or leave a fill unhedged.
-
Under review — #8411: Prevent partial balance snapshots during Bybit Perpetual refresh Proposed constructing refreshed balances in local dictionaries and publishing the completed snapshot atomically, so concurrent readers see either the previous complete state or the new complete state.
-
Under review — #8407: Ignore duplicate stop actions for terminated or shutting-down executors Proposed idempotent stop handling so repeated shutdown actions do not re-enter cleanup logic, raise state errors, or overwrite an executor's existing close reason.
Packet classification, traffic classification, and ML/RL-assisted network systems, with an emphasis on latency, classifier selection, and reproducible performance evaluation.
- Languages: C++, Python
- Systems: Linux, networking, concurrency, performance benchmarking
- Tools: Git, Docker, GitHub Actions, CMake
- ML: PyTorch, scikit-learn
|
|
|
|
|