Embedded Engineer - Machine Learning Systems

About the Company 


At Xailient our vision is a world of seamless and pervasive AI assistance, and our mission is to enable makers and innovators to bring their visions to life by providing the artificial nervous system of a connected world. We are on our way to installing our edge software on 50 Billion devices and we need your help!


Xailient believes that many innovations have yet to come to market, and many existing services are limited in their growth by the costs to create and deliver AI in the real world. Xailient is commercializing breakthrough university research in Artificial Intelligence and Machine Learning from Royal Melbourne Institute of Technology. Our technology dramatically reduces the costs of using Computer Vision AI in real-world applications.


Xailient is a VC funded start-up.

The Technology

Today, the computational effort of AI requires powerful servers, incurring not just cloud costs, but also data transmission, and storage costs. Shipping data to the cloud creates delays and quality-of-service problems. New hardware chips and System-on-Modules seek to provide enough computing power for IoT devices, but the costs are high and these systems draw power and generate heat that makes them unsuitable for many applications.  

Xailient improves the performance of AI by reducing the computation required to extract useful information from real-time video. This improves performance on new and old chips, expanding the market through complementary technology that helps “do more with less” in a hardware-agnostic way.


Xailient’s breakthroughs are inspired by nature. Our CTO-cofounder is a neuroscientist, and we are mapping strategies found in nature into AI. In short, we teach computers to process the way humans think.

The Team

The team is distributed geographically, and our work hours overlap in the Pacific Timezone afternoon to early evening and the Australian Eastern Timezone morning.  We have a daily standup meeting, but otherwise are flexible with work schedules provided team members make time available to collaborate when needed in a way that tries to be fair to all teammates.

The Product Engineering team makes it delightful for customers to use Xailient AI technology.  We are seeking passionate and innovative people to join our Product Engineering team.

We embrace diversity.  Part-time or full-time candidates will be considered.  We welcome candidates with family/life commitments and part-time availability, but “full-time professional passion”.

Role Details

Embedded Machine Learning Engineer


  • Care for Quality - Care about your work and take pride in doing a good job

  • Customer Focus - Desire to serve customers with solutions that delight

  • Communication - Effective communication in written and spoken English

  • Communication - Speak up with questions or concerns, criticize constructively, know when to ask for help

  • Flexibility - Comfortable with changing priorities of a startup environment

  • Flexibility - Excited to learn new things and work outside of technical comfort zone

  • Collaboration - Most ideas, yours and mine, can be improved through collaboration

  • Teamwork - No brilliant jerks, but weirdos welcome

  • Dependability - Deliver on promises, provide status proactively

  • Initiative - see a problem, solve a problem



  • Work closely with research and engineering teams at both Xailient and the customer to facilitate deploying Xailient AI neural networks and models on customer’s chips

  • The ideal candidate should be proficient in using deep learning frameworks (e.g. PyTorch and TensorFlow) to train and deploy deep learning networks to edge devices (e.g. 3rd party ASICs, FPGAs, dedicated AI chips, NPUs, GPUs).

  • Design and compile custom operations in TensorFlow Lite

  • Efficiently implement artificial intelligence (AI) focused perception systems on target hardware

  • Collaborates and adds value through participation in peer code reviews, providing comments and suggestions.

  • Implements, integrates and verifies ML on target hardware accelerators.

  • Troubleshoots to bridge the accuracy gap if needed (Analyzing the errors of the model and designing strategies to overcome them). Performs technical root cause analysis and outlines corrective action for given problems.

  • Must “plumb” and “merge” multiple CNNs together using TensorFlow or PyTorch primitives where required



  • 8-bit, 4-bit, 2-bit, 1-bit quantization

  • Experience cross or native compiling of TensorFlow Lite (or Micro) on embedded devices

  • Strong Deep Learning experience

  • Experience on memory management, memory optimizations, performance measurement and tuning

  • Looking for an expert in machine learning with strong embedded skills and experience.

  • Ability to read and communicate data pipelines via Sequence Diagrams

  • Familiar with classical low-level and fundamental image processing, digital signal processing and computer vision algorithms.

  • 4+ years of experience in embedded ML software design and development.

  • 2+ years of experience with embedded systems for signal processing and machine learning.

  • Must be familiar with generating quantized TensorFlow Lite models. Must be familiar with converting TensorFlow Lite models to ONNX and some custom formats (using 3rd party conversion tools).

  • Ability to design custom TensorFlow or PyTorch layers or find alternative layers if the target embedded devices don’t support certain layers/operations.

  • Good embedded Linux experience

  • Proficiency with OpenCV

  • Comfortable with agile software management and unit testing practices and tools (Git, version control). Work with algorithm experts and software / systems developers in Agile workflow.

  • Fluent in modern programming languages, such as C/C++, Python

  • Multi core/processor system implementation knowledge and experience

  • Experience with ML acceleration – compression, quantization pruning etc.

  • Demonstrates mastery of software engineering tools (configuration management systems, build processes, debuggers, emulators, simulators and logic analyzers).

  • Digital Signal Processing experience

  • Hardware accelerator experience

  • Implement, integrate and verify ML on target hardware accelerators.


Nice to haves (not expected, but let us know):

  • Experience dealing with distributed deployments with versioned interfaces across multiple parties (API wrangling, version management) a plus

  • Computer vision experience

  • Nice to have formal training in one or more cloud platforms

Principals only please; Xailient is not accepting recruiter placements, nor will be responsible for recruitment fees.

or just email us at:  careers@xailient.com