Qiang Fu

About Qiang Fu

Wireless Communications System Design and Optimization (12 years): Led the design, implementation, and optimization of Wireless Communications Systems, specializing in OFDM, PSK, MIMO, LDPC, Channel Estimation, and Equalization.2. Reinforcement learning and Signal Processing (9 years): Experience in machine learning, reinforcement learning, deep RL, RLHF, PyTorch, Gym, signal processing, mathematics, stochastic processes, information theory, and linear algebra.3. Embedded Communications System Development (6 years): Expertise in RTL embedded communications system development, enhancing system efficiency and functionality for Bluetooth and Wi-SUN.4. Bluetooth SIG Core Spec Working Group: Active member contributed to Bluetooth SIG Core SPEC 6.0 development.5. RF Testing (3 years): Proficient in 802.11ax/be and Bluetooth standards; experienced with LitePoint, Ellisys, Keysight, Rohde Schwarz, and Tektronix equipment. Validated DUTs on Linux platforms for BRCM, QCOM, and Infineon.6. Programming Skills (11 years): Proficient in C/C++, MATLAB, and Python.7. Work Authorization: H-1B started on October 1st, 2024, EB-2 NIW(on behalf of myself) Approved.

Experience

1. Cellular System Performance Engineer

   Apple

   Nov 2024 — Present · San Jose, CA, US

   Deep Reinforcement Learning and LLM: 1. Deep Reinforcement Learning with Human Feedback (RLHF): Developed an A2C agent with CNN-based policy and value networks trained on raw pixels, incorporating human preference and scalar feedback for reward modeling to supplement sparse environment rewards. Designed a human-aligned hybrid reward to stabilize policy optimization, achieving faster convergence, improved sample efficiency, and more aligned behaviors compared to standard A2C baselines.2. Deep Reinforcement Learning: Developed a custom OpenAI Gym–compatible stock trading environment and trained Deep Q-Network (DQN) agents in PyTorch using PTAN to maximize long-term returns. Evaluated fully connected and CNN-based DQN architectures with experience replay, target networks, and ε-greedy exploration, analyzing convergence behavior and policy performance across models.3. LLM: Developed a GPT-5.2–powered chatbot web app for domain-specific knowledge-base Q&A using natural language generation. Implemented a JavaScript frontend with real-time API integration, including prompt construction, response parsing, and error handling. Validated and optimized end-to-end inference workflows using OpenAI Playground and Postman to ensure reliable model behavior and robust deployment.Wireless Communications:1. Post-Silicon System Bring-up: Worked on VP and post-silicon bring-up and performance optimization of 5G-NR FR2 cellular chipsets, covering RF chip, IF chip, and system-level integration.2. System Test Development & Validation: Developed and validated system-level test scripts for 5G-NR FR2 chipsets, including 3GPP-defined TX/RX tests and new feature validation.3. System Performance Validation & Analysis: Collected performance data using DARTs and analyzed key system metrics with Tableau to identify and drive performance improvements.

Education

• Nanjing Agricultural University

  Bachelor's degree, Mechanical Engineering

  2009 — 2013

• Xiamen University

  Master of Science (M.S.), Marine Physics

  2013 — 2016

• The University of Alabama

  Doctor of Philosophy - PhD, Electrical and Computer Engineering

  2017 — 2022