Machine Learning Engineer

Machine Learning Engineer

Location: Cape Cod, MA (on-site/hybrid) · Type: Full-time

About Triiodide

Triiodide's mission is to build immersive spaces for friends to spend time together. We operate at the intersection of embodied AI and games. We make best selling games for players to enjoy with friends while also training models to help craft realistic procedural environments, behavioral RL policies to make in-game creatures feel alive, and run inference on edge devices in real-time — the same class of problems faced by frontier labs.

We are a five-person studio behind Backrooms: Escape Together, a co-op survival horror game frequently in the top 100 on Steam and loved by millions.

About the Role

We are looking for a Machine Learning Engineer to work across the full model stack — from reading and implementing papers, to training and evaluating models, to optimizing inference for real-time performance on edge devices (PCs, consoles). You will own problems end-to-end on a team of five.

This is an opportunity to work on the most exciting problems in AI research while having people experience the impact of your work today. Almost everything you build ships to a live playerbase of millions.

What You'll Do • Train generative models that produce simulation/procedural environments — roads, floorplans, terrain using real-world or synthetic data • Train RL-driven locomotion and behavior policies for in-game entities/creatures • Use RLHF with in game telemetry to tailor gameplay and world generation to each player • Optimize ML inference for real-time performance on edge devices (PCs, consoles) — model quantization, ONNX, and strict latency budgeting on consumer hardware • Identify systems that can be replaced or improved with learned approaches

You may be a good fit if you: • Have strong software engineering skills • Have a results-oriented mindset with a bias towards flexibility and impact • Pick up slack, even if it goes outside your job description • Wish to move the world forward with your work, including the frontier of machine learning • Have a strong product sense and desire to win

Strong candidates may also have experience with: • Generative model architectures (diffusion, autoregressive, etc) • Deep learning frameworks (PyTorch/JAX) • Reinforcement learning for robotics (MjLab, Isaac Lab, or similar) • GPU internals • Agentic tools (Cursor, Codex, Claude Code)

The annual compensation range for this role is listed below.

Annual Salary:

$160,000-$225,000 USD (base + bonus)

Compensation Range: $160K - $200K