Key Responsibilities
- Develop post-training, evaluation, and serving systems to transform frontier LLMs into reliable, high-quality product experiences for billions of users
- Design and implement a general-purpose agentic platform to power GenAI products across the ecosystem, enabling faster, safer, and scalable deployments
- Optimize model performance through fine-tuning, post-training, and capacity/cost-efficient techniques for large-scale deployment
- Adapt and scale AI systems across diverse the organization products, ensuring seamless integration and performance
- Lead research in multimodal reasoning and generation, prioritizing solutions with direct product applicability
- Curate high-quality pre/mid/post-training data pipelines for multimodal LLMs and foundation models
Requirements
- Bachelor's degree in Computer Science, Computer Engineering, or equivalent practical experience
- 5+ years of experience as an Applied AI Research Scientist or Engineer
- Hands-on expertise in large-scale model training, algorithm implementation, and evaluation of speech-based systems
- Strong background in post-training pipelines (SFT, RLHF, synthetic data generation) and evaluation methodology
- Experience with production serving systems (RAG, memory, multi-modal generation) and agentic product development