SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Mariana Minerals is a software-first, vertically integrated minerals company supplying critical minerals for modern energy, AI, and defense technologies. The company reimagines the minerals supply chain by combining deep industry expertise with advanced software, automation, and data-driven decision-making.
You will lead a team of machine learning engineers building models for chemistry and process engineering, including surrogate models, hybrid physics-ML models, and optimization/control layers that inform and set how plants and chemical processes operate. This is a player-coach role where you'll manage people, own technical quality, and partner with the Technical Product Manager for ML & Robotics on product direction.
Key responsibilities include:
- Managing, coaching, and growing a team of ML engineers: hiring, onboarding, 1:1s, performance reviews, and career development
- Owning technical direction: model architectures, data strategy, validation methodology, and standards for when models are trusted enough to inform process decisions
- Partnering with the TPM on roadmap: translating priorities into scoped work, pushing back on infeasible asks, and committing to deliverables
- Working directly with process engineers, chemists, and operators to ensure models answer real plant questions
- Setting rigor standards for physics-informed modeling: mass/energy balances, thermodynamic consistency, extrapolation limits, uncertainty quantification
- Owning production lifecycle: deployment, monitoring, retraining, drift detection, incident response, and operational practices (on-call, runbooks, review)
- Running engineering practices: code/model review, experiment tracking, reproducibility, documentation
- Staying hands-on enough to review difficult work, unblock engineers, and prototype when needed
- Owning headcount planning and building a pipeline of ML engineers with chemistry/process backgrounds
You'll operate as a player-coach, spending most time on people, priorities, and quality while staying engaged enough in code and data to maintain sharp judgment. You'll structure ambiguous process problems into scoped modeling work with clear success criteria, apply physics-literate skepticism to validate that models learn chemistry rather than artifacts, provide clear commitments with visible tradeoffs, and understand how the team's models integrate with simulators, data platforms, and control systems.
REQUIREMENTS:
Must have:
- 6+ years in machine learning or scientific computing, including 2+ years managing ML engineers or scientists with direct responsibility for hiring, performance, and growth
- Degree or equivalent depth in chemical engineering, chemistry, materials science, or closely related field; ability to read process flow diagrams, reason about reaction kinetics and separations, and engage with process engineers
- Hands-on track record shipping ML models to production for physical, scientific, or industrial systems: surrogate models, hybrid physics-ML models, time-series forecasting, Bayesian optimization, or similar
- Strong engineering fundamentals: Python, modern ML frameworks, experiment tracking, and discipline to make research code reproducible and maintainable
- Track record of setting technical direction for a team and delivering against commitments in ambiguous, cross-functional environments
- Ability to evaluate work you didn't do: review models, spot validation flaws, and give feedback that improves engineers
- Exceptional written and verbal communication across audiences (ML engineers, process engineers, operators, executives)
Nice to have:
- PhD in chemical engineering, chemistry, or materials science with ML or computational focus
- Experience with model predictive control, real-time optimization, or closed-loop optimization of chemical/industrial processes
- Background in mining, hydrometallurgy, energy, chemicals, or heavy industry
- Experience building a team from a handful of engineers to a functioning organization
