Mariana Minerals is building the critical minerals supply chain from the ground up, and it needs machine learning expertise to push its mineral refining facilities toward autonomy. In this role, you will develop and improve learning-based control systems, starting in simulator-based training pipelines and progressively validating models against real operating plant behavior.

As a Machine Learning Engineer in San Francisco, you will focus on reinforcement learning for autonomous, short-interval control. The goal is to bridge simulation and reality so the system can improve recovery, energy use, reagent consumption, and equipment uptime.

What you’ll do

Run reinforcement learning experiments in physically realistic simulators of mineral processing operations, then help translate results into stronger 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 analyze cases where a model underperforms.
Close the simulation-to-reality gap by comparing model behavior to real plant data and flagging where the physics diverges.
Write clean, well-tested code and contribute to the services that deploy models into production.
Collaborate with process and chemistry experts to understand the unit operations being modeled.

What you bring

2-8+ years of experience (including internships or research) in machine learning, reinforcement learning, or scientific computing, or a strong recent graduate with demonstrated project depth.
Solid grounding in machine learning fundamentals, with 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 a willingness to learn chemistry and process engineering from experts who will challenge assumptions.
A self-starter who asks good questions, ships work, and escalates blockers early.

The tech you’ll use

Python
Reinforcement learning
Deep learning

How the system works

Mariana Minerals’ internal platform applies reinforcement learning toolkits commonly used in self-driving vehicles and humanoid robots to autonomous, short-interval control of mineral refining circuits. Models update operating set points and configurations in real time, optimizing across lithium recovery, reagent consumption, energy intensity, and equipment uptime at the same time.

The environment is noisy and non-stationary, with wastewater compositions that shift, ore grades that change, and equipment aging over time. Training is performed inside physically realistic simulators of process units, followed by validation against real plant data before models are used on live equipment, supporting the end goal of fully autonomous refining operations.

Compensation

Compensation Range: $120K - $180K (USD per year)

Mariana Minerals culture

Extreme Ownership: Taking full responsibility for outcomes and driving relentlessly toward solutions.
Engineer Out Requirements, then Automate: Simplify, optimize, then automate for scale.
Share Your Legos: Collaborate openly, share knowledge, and empower each other to build bigger solutions.

Machine Learning Engineer