Member of Technical Staff - Diffusion Model

Introducing Moonlake, AI for creating world simulations.

About Moonlake

Moonlake is building the frontier of interactive world models: systems that generate, simulate, and reason over 3D environments for embodied AI, robotics and gaming. We develop the simulation infrastructure to build worlds (e.g., assets, scenes, digital twins) at scale.

Our team sits at the intersection of:

• Embodied AI

• Robotics simulation

• Interactive 3D worlds

• World models

• Real-time generation

• AI infrastructure

Moonlake is building the next generation of AI infrastructure for interactive digital worlds. Our mission is to enable anyone to create, simulate, and interact with rich environments using natural language and multimodal inputs, turning simple ideas into worlds with structure, logic, and agents that can perceive and act.

Our team has raised $28M in seed funding from NVIDIA Ventures, Threshold Ventures, AIX ventures and notable angels including Naval Ravikant and Jeff Dean to build the foundational layer for the future of AI - powering everything from creative tools and games to robotics training, simulations, and digital twins. Our goal is to make building and experimenting with these environments as accessible and scalable as publishing video on the internet.

We are looking for exceptional research engineers and applied researchers to help push the frontier of interactive AI.

The Role

We’re looking for a Member of Technical Staff — Diffusion Models to help design and train the next generation of multimodal generative systems powering Moonlake’s interactive world platform.

This is a research-heavy role focused on:

• Diffusion architectures

• Video generation

• Conditioning systems

• Multimodal generation

• Control and personalization

• Large-scale training

The ideal candidate combines:

• Strong ML research fundamentals

• Practical systems intuition

• Experience training generative models at scale

• Deep curiosity around interactive world generation

This role has a very high technical bar. Successful candidates typically have:

• Published research

• Strong generative modeling experience

• Video generation or graphics-related experience

• Prior work on frontier multimodal systems

What You’ll Do

• Build and iterate on diffusion architectures across:

  • 2D

  • 3D

  • Image

  • Video

  • Audio

• Develop conditioning and control systems for multimodal generation

• Improve generation quality, controllability, consistency, and efficiency

• Train large-scale generative models

• Build systems for editing, personalization, and controllable generation

• Collaborate closely with infrastructure, world-modeling, and product teams

• Push generation systems toward real-time and interactive applications

Scope of Work

Modeling & Architecture

• Build and improve diffusion architectures

• Video diffusion systems

• Multimodal generation pipelines

• Latent-space modeling

• Real-time generation architectures

• Interactive generation systems

Conditioning & Multi-Modal Learning

• Text conditioning

• Image conditioning

• Pose/layout/control signals

• Multi-modal encoders

• Guidance strategies

• Structured generation control

Training & Optimization

• Large-scale diffusion training

• Distributed training systems

• Sample quality vs. compute optimization

• Distillation techniques

• Consistency models

• One-step generation systems

• Efficient generation pipelines

Control & Alignment

• ControlNet

• LoRA

• IP-Adapters

• Style / identity / geometry conditioning

• Editing pipelines

• Inpainting systems

• Personalization systems

• DreamBooth and custom tuning workflows

What We’re Looking For

• Strong ML research background

• Deep understanding of diffusion models and generative architectures

• Experience training large-scale generative systems

• Strong grasp of optimization, scaling, and multimodal learning

• Ability to work across both research and implementation

• Strong engineering fundamentals

• Ability to iterate quickly in a fast-moving research environment

Bonus Points

• Experience with 3D generation or world models

• Robotics simulation or embodied AI familiarity

• Interactive generation systems

• Real-time inference optimization

• Graphics or game-engine experience

• Experience building production-grade generation pipelines

Why This Role Matters

Moonlake is not building static image generators.

The company is building systems capable of generating:

• Interactive worlds

• Dynamic simulations

• Controllable environments

• Real-time multimodal experiences

The diffusion stack is foundational to making these systems coherent, controllable, scalable, and interactive.

You’ll help define the generation systems behind the next generation of world-model AI.

We are committed to being an on-site, in-person team currently based in San Francisco.