Machine Learning Engineers at Mariana Minerals help develop and improve machine learning systems that control mineral refining facilities. The role focuses on reinforcement learning for autonomous, short-interval control, progressing from simulator-based training pipelines toward models that operate on real plants, with an emphasis on bridging simulation and reality.
Role Focus
- Develop reinforcement learning capabilities for autonomous control of mineral refining circuits.
- Advance training approaches from physically realistic simulators toward deployment on operating plants.
- Optimize mineral recovery and operational performance while accounting for real-world constraints.
Responsibilities
- Run reinforcement learning experiments in physically realistic simulators of mineral processing operations, then use results to improve controllers.
- Build and refine training environment components, including reward functions, observations, and action logic, with guidance from senior engineers.
- Train control models, track and interpret their performance, and investigate causes when models underperform.
- Reduce the simulation-to-reality gap by comparing model behavior against real plant data and flagging where underlying physics diverge.
- Write clean, well-tested code and contribute to services that place models into production.
- Partner with process and chemistry experts to understand the unit operations being modeled.
Requirements
- 2–8+ years of experience in machine learning, reinforcement learning, or scientific computing, including internships or research, or a strong recent graduate with demonstrated project depth.
- Solid grounding in machine learning fundamentals and working knowledge of modern deep learning; reinforcement learning exposure is a strong plus.
- Proficiency in Python and comfort reading and debugging an existing codebase.
- Curiosity about physical, industrial systems and willingness to learn chemistry and process engineering from experts who challenge assumptions.
- Self-starter mindset, demonstrated ability to ship, and early escalation of blockers.
Technologies
- Python
- Reinforcement learning
- Deep learning
The Tech
- Internal platform uses reinforcement learning toolkits similar to those used in self-driving vehicles and humanoid robots, adapted for autonomous short-interval control of mineral refining circuits.
- Models adjust operating set points and configurations in real time, optimizing across lithium recovery, reagent consumption, energy intensity, and equipment uptime.
- Training occurs in a noisy, non-stationary environment where wastewater compositions shift, ore grades change, and equipment ages, requiring continuous adaptation.
- Training includes closing the gap against real plant data before models touch live equipment.
Compensation
Compensation Range: USD 120,000–180,000 per year.
Location and Employment
- Location: San Francisco, CA (onsite).
- Minimum Experience: 2 years.
Company Context and Impact
Mariana Minerals is building the critical minerals supply chain from the ground up and is working toward autonomous refining operations. The team owns projects, generates data, and closes the loop so that each facility improves software capability and speeds future deployments.
Culture
- Extreme Ownership: Full responsibility for outcomes, focused on driving toward solutions.
- Engineer Out Requirements, then Automate: Simplify, optimize, and automate for scale.
- Share Your Legos: Collaborative knowledge sharing and empowerment to build larger, better solutions.