The Spectacles team is pushing the boundaries of technology to bring people closer together in the real world. Our fifth-generation Spectacles, powered by Snap OS, showcase how standalone, see-through AR glasses make playing, learning, and working better together.

‍

We are looking for a Machine Learning Software Engineer to join our ML tools and technology team at Snap Inc!

‍

‍

What you’ll do:

In this role, you’ll help develop the next generation of on-device intelligence for Spectacles AR glasses. You will not only design and implement cutting-edge ML algorithms, but also build the infrastructure, tools, and workflows that make ML development, deployment, and monitoring at scale possible. Your work will enable seamless, real-time AR experiences, pushing the limits of performance, reliability, and efficiency on diverse hardware platforms. The ideal candidate brings a strong background in software engineering and computer vision, along with hands-on experience developing machine learning optimization algorithms and infrastructure for diverse hardware platforms.

• Design and implement ML workflows and infrastructure for training, fine-tuning, evaluating, and deploying models for AR and on-device applications, with a focus on computer vision and large language models (LLMs).

• Develop and extend ML optimization pipelines, model transformation tools, and runtimes to enable efficient deployment on AR hardware platforms.

• Build and maintain MLOps pipelines for automated training, testing, validation, monitoring, and continuous deployment of ML models.

• Explore and implement advanced model optimizations such as quantization, sparsity and compression techniques, ensuring models meet stringent on-device real-time and power constraints.

• Design benchmarking tools to evaluate correctness, robustness, and performance of ML solutions across hardware and software platforms.

• Collaborate with cross-functional teams to prototype, test, and validate new hardware acceleration approaches, driving them to production.

‍

‍

Knowledge, Skills & Abilities:

• Ability to contribute across the end-to-end lifecycle of machine learning solutions, including design, training, optimization, deployment, testing, and monitoring.

• Strong desire in advancing the internals of ML tooling, such as writing custom operators, improving runtime performance, and building scalable infrastructure for diverse hardware accelerators.

• Proven skill in developing efficient, reliable, and adaptable ML systems that scale across evolving architectures.

• Experience designing scalable training and evaluation systems with a focus on reproducibility and reliability.

• Deep understanding of quality assurance practices to validate ML performance across diverse environments and deployment contexts.

• Capacity to advance team-wide technical maturity by contributing to compilers, SDK integrations, and architectural design that support on-device intelligence at scale.

• Strong communication and collaboration skills, with the ability to align technical innovation with product needs.
  

‍

Minimum Qualifications

• Master’s degree or PhD in Computer Science, Electrical/Computer Engineering, or a related technical field

• 3+ years of professional experience in the field of software engineering.

• 2+ years of experience in testing, deploying, and monitoring production ML systems.

• Proficiency with software development in Python or C++.

• Experience with machine learning frameworks (PyTorch, TensorFlow, etc.) and cloud platforms (GCP, AWS, etc).

‍

‍

Preferred Qualifications:

• Understanding of large language models, NLP and/or multimodal modeling.

• Experience with on-device ML SDKs/tooling (e.g., TensorFlow Lite, ExecuTorch, Core ML, SNPE/QNN).

• Experience in one or more of the following areas: ML performance and efficiency tuning, compiler optimization for ML workloads, hardware-accelerated ML inference, low-level programming models, or distributed ML systems optimization.

• Familiarity with QA automation frameworks and benchmarking at scale.

• Familiarity with the architectural patterns of large-scale software applications.

‍