Hi, I'm Chenyu Hu
EEE Student at Nanyang Technological University, Singapore
Passionate about IC design, signal processing, and large language models & machine learning.
EEE Student at Nanyang Technological University, Singapore
Passionate about IC design, signal processing, and large language models & machine learning.
I'm a second-year undergraduate student studying Electrical & Electronic Engineering at Nanyang Technological University (NTU), Singapore. Currently holding a GPA of 4.25/5.0, I've been offered a place in the Accelerated Bachelor Programme in view of good academic performance.
My academic interests span IC design & VLSI, signal processing, and large language models & machine learning. I enjoy bridging the gap between hardware and intelligent systems — understanding how circuits work at the transistor level while exploring how large-scale AI models process and generate language, vision, and reasoning.
Beyond academics, I'm a Black Belt (1st Dan) martial artist and a competitive Counter-Strike 2 player (Faceit Level 8). Both pursuits have sharpened my discipline, analytical thinking, and ability to perform under pressure — skills that translate directly into engineering problem-solving.
AI / Machine Learning
Designed an evaluation benchmark for long-term memory in emotional companion AI systems. Defined capability dimensions, evaluation tasks, and metrics covering memory updating, forgetting, temporal recall, and persistent user-state tracking.
Embedded Systems / Signal Processing
Built a closed-loop automatic volume control system for an audio amplifier using an STM32 microcontroller (ARM Cortex M4) with MicroPython. Implemented VCA, power amplifier, and VU Meter subsystems on breadboard. Developed Python control algorithms for ADC/DAC data acquisition, FFT spectral analysis, waveform generation, and real-time feedback to maintain constant audio output level.
May 2026 — Present
Robopoet — Shanghai, China
Contributed to the design of an emotional companion AI memory benchmark, supporting the definition of evaluation tasks, capability dimensions, data samples, and metrics for long-term memory, memory updating, forgetting, temporal recall, and persistent user-state tracking. Also evaluated and optimised the overseas AI companion model across multilingual interaction scenarios, supporting memory-system issue detection, Japanese prompt localisation, and schema configuration improvements.
May 2025 — Jul 2025
Hangzhou NetEase Zhiqi Technology Co., Ltd — Hangzhou, China
Annotated customer service dialogue data from e-commerce logistics platforms to support the training of AI-powered customer service models. Processed approximately 200–500 dialogue sessions per day, classifying user intents, service scenarios, and response patterns based on project-specific labelling guidelines.
Aug 2024 — May 2028
Nanyang Technological University — Singapore
GPA: 4.25 / 5.0 | Offered place in Accelerated Bachelor Programme
Sep 2020 — May 2023
Keqiao High School of Zhejiang — Zhejiang, China
Gaokao (National College Entrance Examination) — Top 5% in Zhejiang Province