Forward Deployed Machine Learning Engineer
Stratum AIover 1 year ago
Remote, CanadaMid Level
Responsibilities
- Adapt Stratum’s deep learning models to specific mines.
- Develop and maintain high-quality machine learning code using Python and PyTorch.
- Apply ML to resource and metallurgical modeling for individual mines and mining operations generally.
- Conduct foundational research into accurate AI resource modeling techniques.
- Train and refine models for specific mine operations.
- Track deployed model performance and identify improvements.
- Communicate model quality, metrics, performance, and methodology to external technical stakeholders, geologists, and business development teams.
- Identify additional opportunities to apply Stratum’s technology with existing clients.
- Visit mines at least twice per year and engage with clients on the ground.
Requirements
- At least 2 years of industry machine learning experience.
- Excellent proficiency implementing custom neural network architectures in Python and PyTorch.
- Strong foundational deep learning knowledge, especially with Transformers, convolutional neural networks, and/or graph neural networks.
- Hands-on experience with various neural network types is preferred.
- Ability to work autonomously and take initiative in solving complex problems.
- Strong oral and visual technical communication skills.
- Ability to explain model quality and value to non-machine-learning engineers and mining-industry stakeholders.
Benefits
- Remote-first work arrangement with a preference for applicants based in Canada.
- Opportunity to work on applied and foundational machine learning affecting mining operations and resource efficiency.
- At least two mine visits per year as part of client engagement.
- Ability to grow into a senior engineering role focused on expanding technology use with existing clients.
About Stratum AI
Stratum AI builds AI-driven mineral resource modeling software for mining companies, generating 3D ore body predictions and economic insights from limited geological data. Its web dashboard lets clients upload data, visualize models, and access AI-generated predictions that inform mine planning and reduce waste. Founded in 2018 and headquartered in Toronto, the privately held company is venture-backed by investors including Y Combinator, Founders Fund, Aramco, and Builders VC.