Senior Machine Learning Engineer - Reconstruction

WHO WE ARE:

Peripheral is developing spatial intelligence, starting in live sports and entertainment. Our models generate spatial data, used for advanced sports analytics and immersive media experiences. We're solving key research challenges in 3D computer vision, creating the foundations for the next generation of robotic perception and embodied intelligence.

We're backed by top investors, including Khosla Ventures, Daybreak, and Entrepreneurs First, and working with some of the biggest names in sports. Our team includes engineers and researchers from leading technology companies and research institutions, and we're building technology at the intersection of AI, graphics, and the future of live entertainment. We're ambitious and looking to win.

THE OPPORTUNITY:

We're seeking a Senior Reconstruction Researcher to lead Peripheral's work on spatially intelligent foundation models.

We believe spatial intelligence is one of the next major frontiers in AI, with the potential to transform live sports and media today and accelerate progress in robotics.

At Peripheral, you'll have the ingredients to push this problem forward: large-scale 3D/4D data, the ability to capture more from live deployments, compute, funding to train next-generation foundation models, and a strong research, engineering, and infrastructure team built for rapid experimentation and execution.

You'll work directly with our CTO to set the research agenda and stay deeply hands-on across model architecture, training, datasets, and experimentation. You'll have the freedom to pursue ambitious ideas and the responsibility to turn the best ones into real-world systems by leading a team of talented researchers and infrastructure engineers. You'll also help define how research is conducted at Peripheral, grow the research team, and advance our mission to build spatial intelligence.

WHO YOU ARE:

You've led research teams working on 3D reconstruction, neural rendering, or spatial intelligence, and you know how to turn promising research into systems that ship. You're equally comfortable setting a multi-quarter research agenda and getting into the technical tradeoffs required to bring models into production.

You think deeply about spatial intelligence, from classical multi-view geometry and neural approaches such as NeRFs and Gaussian Splatting to feed-forward 3D reconstruction and world models, and you have strong views on where the field is headed next.

You're a multiplier for the people around you. You know how to give researchers room to explore while providing engineers with clear direction, and you can connect research and production so both move faster.

You're excited to shape Peripheral's research culture from the ground up, setting the technical bar, rigor, and ways of working the team will build on as it grows. Most importantly, you're motivated by the frontier we're working toward and aligned with our mission to build models that understand the spatial world around them.

WHAT YOU'LL BE DOING:

• Set and drive Peripheral's research agenda in spatial intelligence, including the development of models for real-time, feed-forward photorealistic 3D and 4D reconstruction.

• Lead and grow the research team, setting technical direction in collaboration with the CTO from early exploration through production.

• Stay deeply hands-on with model architecture, novel training strategies, experiment design, model evaluation, dataset strategy, and production deployment.

• Evaluate emerging research from academia and industry to determine which are worth pursuing.

• Advance Peripheral's research by contributing to publications, patents, and technical talks where appropriate.

REQUIREMENTS

• Master's or PhD in Computer Science, Robotics, Computer Vision, Machine Learning, or a related field.

• Proven track record leading research in 3D reconstruction, neural rendering, or spatial intelligence from prototype to a live, production-grade product.

• Publications at top-tier venues such as CVPR, ICCV, ECCV, NeurIPS, or 3DV, and/or a strong record of patents or open-source contributions.

• Deep expertise in multi-view geometry, structure-from-motion, NeRFs, 3D/4D Gaussian Splatting, and modern feed-forward 3D reconstruction.

• Strong hands-on experience designing model architectures, training strategies, rigorous experiments, and evaluation protocols.

• Experience with managing data storage, annotation, versioning, and growth of data pipelines, including processing multi-view video, LiDAR data and calibration.

• Experience with distributed PyTorch training using DDP, FSDP, or similar approaches across multi-node GPU clusters.

• Experience training ML models in cloud or HPC environments across multi-node GPU clusters.

• Experience leading researchers and engineers, setting clear technical direction, and driving execution against research priorities.

• Must have the legal right to work in Canada and be willing to relocate to Toronto. We are unable to provide immigration sponsorship.
  

NICE TO HAVE:

• Background in robotics, autonomous systems, world models, diffusion models, or related areas of spatial intelligence.

• Strong understanding of the NVIDIA ecosystem, including CUDA and GPU architecture for performance optimization, NVDEC/NVENC for video processing, and Isaac Sim/Omniverse for synthetic data generation.

• Experience training foundation models, with a strong understanding of scaling laws, compute economics, and how to prioritize high-value research bets.

WHY YOU'LL LOVE WORKING HERE

• Competitive equity package as an early team member.

• Annual salary of $250K-$300K CAD plus performance bonuses, commensurate with experience.

• Unparalleled access to premier global sporting events and iconic venues.

• Full ownership of high-impact projects shaping the future of spatial intelligence and 3D media.

• Flexible Paid Time Off (PTO).

• Comprehensive health, dental, vision, and wellness benefits.