Senior Machine Learning Engineer, Reinforcement Learning – Egofold

Snail Games USA•Beverly Hills, CA

12h•$150,000 - $185,000•Hybrid

About The Position

Egofold is an AI initiative within Snail Games focused on intelligent agents, simulation, and AI-driven workflows for interactive products. It operates with startup-style speed and broad ownership, backed by an established game company, and is currently building practical prototypes while shaping its longer-term direction. We are looking for a Senior Machine Learning Engineer with strong depth in machine learning and practical experience applying reinforcement learning and related methods to agent behavior and decision systems. This role is focused on the ML core of Egofold: designing experiments, training and improving models, shaping evaluation loops, and helping successful approaches become usable parts of the broader project. This is not a siloed research role. The best candidates stay engaged through evaluation, iteration, and practical integration, and bring enough adjacent breadth to be effective in a small, collaborative team. We value curiosity, ownership, sound judgment, and respectful, low-ego collaboration.

Requirements

Strong foundation in machine learning, with hands-on experience building, training, and iterating on applied ML systems.
Professional or substantial project experience with reinforcement learning, agent-based systems, sequential decision-making, or closely related areas.
Strong Python skills and experience with modern ML frameworks such as PyTorch.
Experience designing experiments, evaluating model behavior, and improving results through systematic iteration.
T-shaped capability: deep machine learning expertise plus practical range across one or more adjacent areas such as simulation, evaluation, model integration, systems collaboration, or robotics-adjacent machine learning.
Strong problem-solving ability, sound judgment, and comfort working in ambiguous, fast-changing environments.
Respectful, low-ego collaborative style and willingness to work beyond a narrow specialty when the work requires it.

Nice To Haves

Experience with reinforcement learning methods such as PPO, SAC, DQN, actor-critic, or related approaches.
Familiarity with simulation environments, multi-agent systems, game AI, or interactive agent behaviors.
Familiarity with C++, inference runtimes, or collaborating with engineers who deploy machine learning models into production systems.
Exposure to robotics, embodied AI, or embedded / on-device machine learning constraints.

Responsibilities

Design, train, and iterate on machine learning models for intelligent agents and decision-making systems, with an emphasis on reinforcement learning and related approaches.
Define and refine state representations, action spaces, reward structures, and evaluation criteria to improve agent behavior.
Build and improve practical experimentation and training workflows, including data generation, experiment tracking, and reproducibility.
Analyze results, debug model behavior, and make pragmatic tradeoffs between model performance, iteration speed, and system complexity.
Work closely with engineers and other partners to help integrate successful ML work into usable product systems.
Contribute thoughtful technical input on next-step experiments, tooling, and ML direction as Egofold continues to evolve.