E

EdgeCortix

Fabless AI accelerator designer for edge

Machine Learning Engineer

Applied Machine LearningConfirmed live in the last 24 hours
Full-timeKawasaki, Kanagawa, JapanIn-person

About the Job

Requirements

  • Experience writing Python code.
  • Experience using TensorFlow and PyTorch machine learning frameworks.
  • Strong knowledge of deep neural networks.
  • Strong understanding of network implementation, training, and accuracy evaluation techniques.
  • Strong debugging and analysis skills for root-causing complex issues.
  • Experience with Git and GitHub or GitLab development workflows.

Responsibilities

  • Work closely with compiler and hardware engineers to accelerate adoption of state-of-the-art neural networks on the company's proprietary hardware and software stack.
  • Guide deployment of custom and off-the-shelf models to meet customer AI system accuracy and performance targets.
  • Grow and maintain a collection of ready-to-use models for perception tasks.
  • Work with the compiler team to improve software-stack stability and user experience.
  • Work with existing neural network implementations, implement models from scratch, train, quantize, prune, and deploy models on company hardware, conduct accuracy studies, review academic literature, and work with public and proprietary vision datasets.
  • Communicate with hardware and compiler teams.

Desired Qualifications

  • Strong object-oriented design and development skills.
  • Knowledge of state-of-the-art neural network architectures for perception tasks such as object detection, semantic segmentation, and monocular depth estimation, and of popular vision datasets such as COCO and KITTI.
  • Experience using deep learning compiler frameworks such as TVM, Glow, or XLA.
  • Experience deploying models and developing machine learning applications on embedded devices.
  • Experience with network quantization and pruning techniques.
  • Experience customizing network implementations and tuning hyperparameters.
  • Good knowledge of recurrent neural networks, long short-term memory networks, and Transformer architectures; experience with point-cloud processing, sequence networks such as natural language processing, and graph neural networks.
  • Knowledge of neural architecture search techniques.
  • Knowledge of reinforcement learning.

About the Company

EdgeCortix

Fabless AI accelerator designer for edge

EdgeCortix designs energy-efficient AI accelerators and software for edge computing in a fabless model, co-designing hardware and software to optimize AI performance on edge devices. Its main products are AI-specific processors and a software platform that, together, enable deploying AI at the edge with low power consumption. Unlike typical hardware vendors, EdgeCortix emphasizes software-centric co-design to maximize efficiency, aiming to deliver scalable edge intelligence for sectors like defense, robotics, and automotive. The company’s goal is to enable practical, power-conscious AI processing on edge devices, supporting broad deployment through its integrated hardware-software ecosystem and ongoing growth funded by investors.

Company size
51-200
Company stage
Debt Financing
Total funding
$197M
Headquarters
Tokyo, Japan
Founded
2019

Recently Posted Jobs

Explore more opportunities at EdgeCortix

See more jobs