I'm Wang Haolin, an AI major at Sichuan University (SCU) in Chengdu. I build LLM agents and the harnesses that drive them, and push embodied AI from simulation onto real legged robots and robot arms — with a side track in RL for high-frequency markets.

• 🎓 Artificial Intelligence @ Sichuan University · Chengdu, China
• 🤖 LLM agents & harnesses — multi-agent pipelines, MCP tooling, skill systems, context engineering
• 🦿 Embodied AI / VLA — RL locomotion (PPO, imitation, AMP), π0.5 VLA deployment on edge devices
• 📈 RL for markets — market making & microstructure-based direction prediction at tick level
• 🧠 Methods I use daily: PPO · imitation learning · domain randomization · teacher-student distillation · sim2real
• 📫 Reach me: whl@scu.edu.cn

• π0.5 VLA real-robot deployment — LoRA-finetuned π0.5 (PaliGemma 2B + action expert) running closed-loop on NVIDIA Jetson Thor × AgileX Piper arm: natural-language pick-and-place with 10-step action chunks, full perception → inference → control loop
• DreamWaQ for wheel-legged robots — adapted DreamWaQ (PPO + β-VAE implicit terrain imagination) to 4×4 wheel-legged platforms: up to 1 m step climbing, Isaac Gym → ONNX → MuJoCo sim2sim → C++ deployment on real robots
• Backflip on a 60 kg quadruped — DeepMimic-style imitation on ZJ-T30-V2: trajectory-optimization reference + PPO tracking, phase-driven control, motor torque-speed envelope
• AMP & CAMP multi-gait control — adversarial motion priors for natural trot, rough-terrain curriculum and fall-recovery; skill-conditioned CAMP learns walk / trot / pronk / bound in a single policy
• Extreme Parkour — end-to-end depth-vision parkour on Unitree Go2: teacher-student privileged RL, depth-latent + heading distillation, ROS 2 deployment
• HIMLoco for Go2W — human-imitation locomotion (ICLR 2024) re-implemented on the wheel-legged Go2W: Isaac Gym training + MuJoCo sim2sim

• AI by Hand (Excel) — deep learning rebuilt from scratch in spreadsheets: backprop, RNN/LSTM/xLSTM, ResNet, full-stack Transformer, Mamba, all the way to AlphaFold — the best way to really understand the math

• RLMM — reinforcement-learning market making on Binance ETH-USDT perps: tick-level hftbacktest Gym environment, RecurrentPPO + LSTM with a 4-phase curriculum over ~900 days of L2 orderbook data; plus a 66-dim microstructure direction model (84.5% validation accuracy, profitable on 93/100 out-of-sample days)

• 🤖 Agent harness engineering — long-running loops, context budgeting, skill ecosystems
• 🧠 VLA on the edge — shrinking π0-class models onto Jetson-class devices with usable control rates
• 🦿 Wheel-legged sim-to-real — taking Go2W RL policies from Isaac Gym onto the real robot
• 🏞️ Agile skills — imitation & contrastive learning for parkour-class maneuvers

_{⚡ "知行合一 — knowledge and action as one." ⚡}