We are looking for a Senior Applied Scientist to help drive the research and development of real-time multimodal conversational AI. You will contribute across two focus areas: advancing foundation models for speech and audio, and building the post-training systems (reward modeling, reinforcement learning) that shape natural, human-like conversational behavior.
You will own a significant research area and contribute across the full model lifecycle - from pre-training and architecture design through post-training alignment and real-time deployment. You will work at the frontier of what's possible in conversational AI, with the compute, data, and runway to pursue problems that few teams in the world have the resources to tackle.
As a Senior Scientist, you will drive the technical execution of your research area, contribute to the team's roadmap, and work closely with inference engineers to ensure your models are designed for real-time production deployment.
Key job responsibilities
What You'll Do
Foundation Model Scaling
- Help build and train large-scale multimodal foundation models for real-time speech and audio generation, from architecture design through production-scale training
- Advance the scaling and efficiency of conversational models, including the relationship between data, model size, and real-time performance
- Design model architectures informed by hardware constraints and inference requirements, working with inference engineers to ensure models are servable from inception
- Develop training methodologies for multimodal models that jointly process and generate speech, language, and audio in real-time streaming contexts
- Contribute to the state of the art on efficient architectures and training methods for conversational AI at scale
Post-Training & Reinforcement Learning
- Design and build reward models and reward functions for speech systems - capturing naturalness, fluency, conversational quality, and real-time responsiveness
- Develop and apply reinforcement learning methods to shape conversational behavior - teaching models natural timing, responsiveness, and fluid interaction
- Build parts of the post-training pipeline from SFT through RL alignment, optimized for real-time multimodal outputs rather than text-only generation
- Design evaluation frameworks that capture the quality dimensions unique to real-time conversation (latency sensitivity, audio quality, prosody, interaction naturalness)
Real-Time Perception & Generation
- Advance the team's capabilities in real-time perception - the ability of the model to process incoming audio/speech while simultaneously generating responses
- Develop techniques for natural interactive systems where the model handles concurrent input and output with human-like timing
- Work at the intersection of model architecture and production constraints to ensure multimodal capabilities function within hard real-time latency budgets
BASIC QUALIFICATIONS
- 5+ years of building machine learning models for business application experience
- PhD, or Master's degree and 6+ years of applied research experience
- Experience programming in Java, C++, Python or related language
- Hands-on experience training large-scale foundation models - direct involvement in model training, not just using pre-trained models
- Experience with multimodal model architectures that jointly process or generate across speech, text, and audio modalities
- Strong understanding of transformer architectures and their application to speech/audio domains
- Publication record at top-tier venues (NeurIPS, ICML, ICLR, Interspeech, ICASSP, ACL, or equivalent)
- Demonstrated ownership of a research area - driving technical direction for a workstream and collaborating effectively across scientists and engineers
PREFERRED QUALIFICATIONS
- Hands-on experience building real-time AI systems - speech, audio, or video
- Track record with post-training methods: reinforcement learning, reward modeling, RLHF/RLAIF, or alignment techniques applied to generative models
- Experience with real-time interactive systems - models that handle concurrent input and output
- Experience building speech-to-speech or audio-to-audio generative models (codec models, autoregressive audio generation)
- Hands-on design of reward models or reward functions specifically for speech/audio quality, naturalness, or conversational behavior
- Experience with distributed training at scale - including parallelism strategies, training stability, and curriculum design
- Familiarity with hardware-informed model design - understanding how architecture choices affect inference latency, memory, and cost
- Background in speech recognition, speech synthesis (TTS), or speech enhancement
- Experience shipping research to production at scale - models serving real users, not just benchmark results
- Contributions to open-source speech/audio ML systems or widely used research codebases
- Experience mentoring junior scientists and engineers
The base salary range for this position is listed below. Your Amazon package will include sign-on payments and restricted stock units (RSUs). Final compensation will be determined based on factors including experience, qualifications, and location. Amazon also offers comprehensive benefits including health insurance (medical, dental, vision, prescription, Basic Life & AD&D insurance and option for Supplemental life plans, EAP, Mental Health Support, Medical Advice Line, Flexible Spending Accounts, Adoption and Surrogacy Reimbursement coverage), 401(k) matching, paid time off, and parental leave. Learn more about our benefits at https://amazon.jobs/en/benefits.
USA, CA, Sunnyvale - 192,200.00 - 260,000.00 USD annually
USA, MA, Boston - 167,100.00 - 226,100.00 USD annually
USA, WA, BELLEVUE - 167,100.00 - 226,100.00 USD annually
USA, WA, Seattle - 167,100.00 - 226,100.00 USD annually
