PaXini Tech Tech Stack

Haptic sensor technology and embodied AI for robotics and automation

Consumer Electronics Shenzhen, Guangdong 501–1,000 employees Privately Held

PaXini Tech develops multi-dimensional tactile sensing hardware and embodied AI algorithms, with a tech stack built around simulation-to-reality (sim2real) workflows: PyTorch, TensorFlow, ROS, MuJoCo, and NeRF for real2sim reconstruction. The engineering-heavy org (15 of 21 roles) is actively tackling sim2real integration and large-scale RL training—the core technical barriers to deploying learned behaviors on physical hardware. Active projects span real robot deployment, digital twin algorithms, and sensor fusion, suggesting the company is moving from research prototypes toward production robotics systems.

Tech Stack 23 technologies

Core StackC++ Python MATLAB PyTorch TensorFlow Rust Linux Unity ROS C/C++ C NeRF Mujoco PyBullet V-REP MuJoCo Lua Bash CUDA Triton Warp Jax

What PaXini Tech Is Building

◆Challenges

Optimizing algorithm performance
Sim2real integration
Large-scale rl training
Real robot deployment
Optimizing internal processes
Optimizing sim2real transfer
Improving algorithm robustness
Organizational change
Recruitment cycle

▲Active Projects

Real2sim scenario reconstruction
Real robot deployment
Numerical solver development
Embodied ai algorithm development
Numerical method research
Digital twin algorithm development
Reinforcement learning deployment
Sensor fusion algorithm development
Sensor math model development
Data collection pipeline

Hiring Activity

Steady20 roles · 9 in 30d

Department

Engineering

Research

Design

Executive

Ops

Seniority

Mid

Senior

Director

Intern

Junior

Notable leadership hires: Algorithm Director, HR Director

Company intelligence

Find more companies like PaXini Tech by tech stack, pain points and active projects

Get started free

About PaXini Tech

PaXini Tech is a Shenzhen-based developer of haptic and tactile sensing technology, positioning itself as China's first to deploy multi-dimensional, multi-array tactile sensors at scale. The company builds both hardware (tactile sensors) and software (embodied AI algorithms and digital twins) for robotics, automotive, wearables, healthcare, and industrial automation applications. With 501–1,000 employees and a research-informed engineering organization, PaXini operates across algorithm development, real robot testing, and sensor hardware integration. The founding team includes alumni from Waseda, Tsinghua, and University of Illinois Urbana-Champaign, with prior experience at ZTE, Xiaomi, and NetEase.

HeadquartersShenzhen, Guangdong

Company Size501–1,000 employees

Hiring MarketsChina

Frequently Asked Questions

What is PaXini Tech's tech stack?

Core languages: C++, Python, Rust. ML/simulation: PyTorch, TensorFlow, MuJoCo, NeRF, PyBullet. Robotics: ROS. Compute: CUDA, Triton. Also uses MATLAB, Unity, V-REP, Lua, Linux, Bash, Jax, and Warp.

What is PaXini Tech working on?

Active projects include real robot deployment, embodied AI algorithm development, sim2real transfer, digital twin algorithms, sensor fusion, real2sim reconstruction, and reinforcement learning deployment. Key challenge areas: sim2real integration and large-scale RL training.