26 days ago
Base Salary
$248k - $310k/yr
Responsibilities
- Fine-tune state-of-the-art LLMs and optimize them for high-performance deployment on Airbnb’s ML infrastructure.
- Build, prototype, and productionize LLM models and pipelines at scale for batch and real-time use cases.
- Work with large-scale structured and unstructured data to develop and continuously improve foundation models.
- Evaluate foundational models and select or refine them for quality, performance, and efficiency.
- Create a multi-year technical roadmap for the team and drive key AI architectural decisions.
- Contribute to Airbnb’s ML platform architecture and strategy while partnering with product, engineering, data science, and operations teams.
Requirements
- PhD in Computer Science, Machine Learning, Mathematics, Statistics, or a related technical field.
- 10+ years of experience developing machine learning models and products at scale from inception through business impact.
- Programming experience in Python and hands-on experience with frameworks such as PyTorch.
- Demonstrated experience training, fine-tuning, optimizing, and improving model inference runtimes.
- Post-training experience in areas such as fine-tuning data processing, responsible LLMs, LLM alignment, reinforcement learning, efficient training and inference, language model evaluation, or multilingual and multimodal modeling.
- Specialized experience in runtime optimization, model quantization, compression, on-device inference, GPU inference, PyTorch, or kernel development.
Benefits
- US remote-eligible work arrangement, with possible occasional Airbnb office or offsite attendance.
- May be eligible for bonus, equity, benefits, and Employee Travel Credits.
Categories
AI ResearchML Engineering
About Airbnb
Airbnb operates a global online marketplace that connects hosts offering homes, rooms, and unique properties with travelers seeking short-term stays and experiences. The company earns revenue primarily through service fees on bookings; founded in 2007 and headquartered in San Francisco, it is a public company on NASDAQ (ABNB) with over 5 million hosts and guest arrivals recorded in nearly every country.