xAI
Software Engineer - Voice Model
About this role
Join SpaceXAI's Grok Voice Model team to engineer cutting-edge voice AI systems that deliver natural, low-latency multilingual conversations. You'll own the full pipeline from data curation and audio processing through model training, evaluation, and real-time deployment at scale.
What you'll do
- Design and execute large-scale speech data curation, collection, synthetic generation, and automated annotation workflows
- Develop pre-training and post-training strategies for speech-language models using supervised fine-tuning and reinforcement learning
- Build comprehensive evaluation frameworks with objective metrics, human preference studies, and A/B testing infrastructure
- Integrate voice models into production applications and real-time environments with low-latency, stable performance
- Collaborate with product teams to define spoken interaction specifications and drive quality improvements
- Optimize models for accuracy, factuality, natural expressiveness, and multilingual fluency
What they're looking for
- Python (deep proficiency)
- JAX or PyTorch
- Large-scale distributed training systems and Kubernetes
- Data processing with Spark and Ray
- Speech-language model pre-training and post-training
- Evaluation pipeline design and metrics
- Reinforcement learning
- Machine learning systems architecture
Benefits
- Competitive base salary with equity
- Comprehensive medical, vision, and dental coverage
- 401(k) retirement plan
- Short and long-term disability insurance
- Life insurance
- Various discounts and perks
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
xAI
xAI builds advanced AI infrastructure and systems, including the Grok model inference platform and Colossus GPU cluster. The company is hiring Mechanical, Electrical, and Facilities Engineers to design and maintain its data center operations, as well as Software Engineers to optimize high-performance inference systems and datacenter networking.
View all jobs at xAILikely interview questions
- Walk us through your experience building or optimizing large-scale data pipelines for speech or audio processing—what tools and techniques did you use?
- Describe a time you worked on pre-training or fine-tuning a language or speech model. What evaluation metrics did you track, and how did you iterate on quality?