AI Engineer, Specialist - VR
Job Description
Build and deploy AI that powers immersive VR and other interactive experiences at Meta. This onsite role in New York, NY focuses on large-scale machine learning systems, spanning deep learning, large language models, recommendation, and generative AI. You will own model quality, reliability, and production deployment at scale, with a strong emphasis on evaluation, monitoring, and safe rollout practices.
Compensation: USD 121,992 to 181,000 per year. Experience: 4+ years.
What you’ll do
- Design, develop, and deploy large-scale machine learning models and AI pipelines across Meta product surfaces, including ranking, recommendation, and generative AI systems
- Own end-to-end technical design for AI/ML systems, evaluating architectural trade-offs and selecting modeling approaches, training strategies, and inference frameworks
- Instrument and profile ML training and serving pipelines to identify bottlenecks, reduce latency, and optimize compute efficiency at scale
- Establish and maintain model evaluation frameworks, including offline metrics, online experimentation, and production monitoring for quality and reliability
- Partner with product managers, data scientists, and research scientists to translate product requirements into ML problem formulations and deliver measurable improvements to key metrics
- Improve AI tooling and infrastructure by building reusable components, strengthening developer workflows, and driving adoption of model development and deployment best practices
- Apply privacy-by-design principles and integrity safeguards across the ML development lifecycle while meeting data governance and compliance requirements
- Mentor engineers on ML system design, modeling techniques, and engineering craft, and support recruiting and technical community programs
- Use AI-augmented development workflows to broaden scope across data analysis, product thinking, and operational readiness beyond core modeling
- Lead staged model rollouts using feature flags and A/B testing frameworks, monitor for regressions, and run incident retrospectives to prevent recurrence
Required experience
- 4+ years of experience designing, training, and deploying machine learning or deep learning models in production environments
- Experience with ML frameworks such as PyTorch or TensorFlow, and strong proficiency in Python for model development and data processing
- Experience with large-scale distributed training, model optimization techniques (such as quantization, distillation, and pruning), and ML serving infrastructure
- Experience designing and running online experiments (A/B testing) to evaluate impact on product metrics
- Ability to communicate technical decisions and trade-offs in writing for both technical and non-technical stakeholders, including design documents and postmortems
Preferred qualifications
- Track record of driving cross-team technical initiatives that improved ML platform capabilities or established new modeling standards
- Experience with large language models, generative AI systems, or foundation model fine-tuning and alignment techniques
- Experience building recommendation or ranking systems at internet scale, including feature engineering, retrieval, and scoring pipelines
- Experience developing ML observability tooling, such as dashboards, alerting, and automated anomaly detection for production model health
Technologies
PyTorch, TensorFlow, Python