Join Our Core AI Model Optimization Team at Meta

Meta's Core AI Model Optimization and Research (MORE) team is on the lookout for talented Software Engineers (SWEs) to enhance our Model Optimization and Training Frameworks. If you're passionate about artificial intelligence and looking to make a significant impact in ML technologies, this position might be perfect for you.

About the MORE Team

The MORE team operates under the XRTech umbrella and is organized into several critical areas:

1. Training Frameworks and Training Efficiency
2. Model Optimization for On-Device Inference
3. Foundational On-Device Models

We focus on powering on-device models across our Family of Apps as well as VR and AR technologies to bring AI functionalities directly to users. Our significant contributions include deploying on-device SAM on IG and pioneering Full-body Avatars on Quest-3. We're also responsible for developing and managing the Vizard training framework extensively utilized across XRTech for optimal model development and enhancement.

Job Responsibilities:

• Optimize AI models to reduce latency and power consumption across on-device and GPU platforms.
• Collaborate with partner teams to navigate constraints related to quality, latency, and complexity in resource-limited environments.
• Develop and enhance tools for automating model optimization and compression.
• Implement fine-tuning, quantization, and deployment of models on mobile and AR/VR devices.
• Advance our newly established Vizard training framework by ideating and rolling out innovative features aimed at improving the developer experience and operational efficiency.
• Spearhead adoption of the framework by integrating key training workloads and providing support for ongoing transitions.
• Offer periodic on-call support and address user issues to streamline model operations.
• Work alongside the Vizard team to elevate the framework’s reliability and efficiency.

Minimum Qualifications:

Candidates should either currently possess or be in the process of obtaining a Bachelor’s degree in Computer Science, Computer Engineering, or a related technical field. It is required that degrees be completed prior to commencement at Meta. Additionally, you should have:

• Specialized experience in model quantization, compression, on-device inference, GPU inference, and utilization of PyTorch.

Preferred Qualifications:

• Advanced degree (Masters/PhD) in Computer Science, Computer Engineering, or relevant field.
• Proven experience in building ML frameworks like Pytorch and Pytorch Lightning.
• Expertise in managing distributed systems and optimizing resource employment.
• Proficiency in accelerating deep learning models particularly for on-device applications.
• Familiarity with on-device inference platforms such as ARM and Qualcomm DSP.
• Experience in optimizing ML model inference and training on NVIDIA GPUs.

Why Choose Meta?

At Meta, we build technologies that enable people worldwide to connect, find communities, and grow businesses. Starting with the launch of Facebook in 2004 through to our advances in the realms of VR and AR, we strive to pioneer the future of social technology. Join us in moving beyond traditional digital connections and help shape the digital future.

Meta values diversity and inclusivity and is committed to providing reasonable accommodations in our recruitment processes. Learn more about the compensation, benefits, and unique opportunities to build a fulfilling career with us.

Ready to advance your career in AI and machine learning? Apply now to become a Software Engineer in SystemsML, Training Frameworks, and Inference Optimization at Meta and help us transform the future of tech.

Compensation:

Annual salary range: $117,000 to $173,000, plus bonuses, equity, and comprehensive benefits based on skills, qualifications, and experience.

Meta is an Equal Opportunity Employer.

Location of Employment:

Flexible / Remote

Contact:

To apply or request accommodations, please email [email protected].