Research Engineer - World Models

Skyfall AI; Toronto; On-site; CAD 90,000 - 150,000; Full time

Job summary

Skyfall AI is seeking a Research Engineer (ML) to join our AI research team in Toronto, Canada. You will develop scalable training infrastructure for world models, work on reinforcement learning, and contribute to open-source projects.

You will collaborate with researchers to implement state-of-the-art architectures like JEPA and transformers, optimize training pipelines, and publish high-impact work for the research community.

Qualifications

• Bachelor's degree in Computer Science, ML or related field.
• Strong Python programming and software engineering practices.
• Experience with cloud-based GPU training environments (AWS, GCP).
• Experience with PyTorch, TensorFlow, or JAX.
• Experience with large-scale distributed systems and training pipelines.

Responsibilities

• Develop scalable AI training pipelines for world models.
• Implement state-of-the-art architectures like JEPA and transformers.
• Optimize training performance and inference for enterprise apps.
• Publish research and contribute to open-source projects.
• Work with cloud-based GPU environments and distributed systems.

Skills

Python

Distributed systems

Research mindset

Software engineering

Cloud GPUs

Education

Bachelor's degree in CS/ML

Master’s degree

Tools

PyTorch

TensorFlow

JAX

AWS

GCP

Job description

Skyfall is building the first enterprise-scale World Model.

Our goal is to build a latent world model that gives agents a human-like sense of foresight in complex digital environments. Unlike traditional physical world models for robotics or autonomous driving, which focus on geometry, physics, and low-level control, our model focuses on semantic and predictive structure in tasks such as navigating enterprise software, booking flights, or operating online stores.

We're looking for a Research Engineer (ML) to join our cutting-edge AI research team. This role is ideal for engineers who thrive at the intersection of AI research and scalable software engineering, working on next-generation world models, reinforcement learning, and multi-agent systems. You’ll play a key role in developing AI training infrastructure for world models, and contributing to the broader research community through publications and open-source projects.

Key Responsibilities:

• Develop Scalable AI Infrastructure – Design and build high-performance training pipelines for world models, multi-modal latent representations, and multi-agent systems.
• Implement Cutting-Edge AI Techniques – Work with state-of-the-art architectures, including JEPA, transformer models and diffusion models.
• Optimize AI Model Performance – Collaborate with researchers to improve training efficiency, fine-tuning strategies, and inference optimization for real-world enterprise applications.
• Contribute to Research & Open Source – Publish high-impact research, engage with the broader AI community, and contribute to leading open-source AI projects.
• Work with Large-Scale Systems – Leverage cloud-based GPU environments and distributed computing frameworks to train and deploy large-scale AI models.

Minimum Qualifications:

• Bachelor's degree in Computer Science, Machine Learning, or a related technical field.
• Strong programming skills in Python, with experience in software engineering best practices.
• Experience with cloud-based GPU training environments (e.g., AWS, Lambda Labs, GCP).
• Hands-on experience with open-source AI frameworks (e.g., PyTorch, TensorFlow, JAX).
• Experience working with large-scale distributed systems and training pipelines.

Nice to Have Qualifications:

• Master’s degree in Computer Science, Machine Learning, or a related technical field.
• Published research in top AI/ML conferences (e.g., NeurIPS, ICML, ICLR, ACL).
• Hands-on experience in LLMs, reinforcement learning, or multi-agent systems.
• Experience optimizing training pipelines for large-scale AI models.
• Contributions to open-source AI projects or AI research communities.