Physics-informed Machine Learning Engineer/Researcher

Physics-informed Machine Learning Engineer/Researcher

Department
Research
Location
HCMCSeattleRemote

Koidra is hiring top-notch machine learning and computational engineers for highly specialized research directions that fuse Physics and Machine Learning with global application impact. Topics that we're dealing with on a regular basis include Graph Neural Network, Neural Differential Equations or Physics-informed Deep Learning, Model-based Reinforcement Learning, etc.

About Koidra

👉 Bio: AIoT company based in Seattle, founded by an ex-Microsoft researcher and backed by top VCs

👉 Mission

is to transform high-impact industries, … enabling companies to meet sustainability and production goals … by automating AIoT-powered digital decisions

We combine years of world-class machine learning research, industrial automation, and operational management to make products that break through previous barriers of real-time automation. We enable manufacturing companies making better digital decisions and consequentially help them become more efficient and more relevant in the future.

👉 Success: repeated global awards and research grants for its game-changing tech

Koidra’s cross-disciplinary AIoT technology
Koidra’s cross-disciplinary AIoT technology

Read About Us for more information about Koidra, its mission, vision, and technologies.

Responsibilities

1️⃣ You will push the state of our proprietary physics-informed ML framework(s) to the next level

2️⃣ You will develop and train end-to-end models

3️⃣ You will develop AI for Good → You won’t target ads, optimize signup funnels, or recommend anything for anyone to click on.

Qualifications

Must Have

✅ Excellent programming, debugging, and analytical skills.

Our team loves writing clean, bug-free, and maintainable mathematical code. We take teamwork to the extreme and hence even for research projects, we still place a high bar for code clarity and robustness.

✅ Experience in constructing and training sophisticated Deep Learning models

✅ Solid understanding of mathematical modeling, differential equations, numerical methods, matrix computations

Preferred Qualifications

✅ Background in computational mathematics/physics/chemistry, scientific computing, control theory, or a related area

✅ Experience with Deep Learning fundamentals. We value candidates who

  1. are able to construct new models (as opposed to using stock CV/NLP models);
  2. understand the hierarchies and computational flow through each building block.

✅ Experience in building models for time series or sequential data (e.g. in NLP) ✅ Familiarity with and strong interests in one or more of the following machine learning research areas: Physics-informed Deep Learning, Graph Neural Network, Model-based Reinforcement Learning, Uncertainty Quantification