Staff Software Engineer, Machine Learning

As a Staff Software Engineer on the ML Platform Team at WellSaid Labs, you will play a crucial role in elevating and extending our machine learning platform. Your responsibilities will include optimizing the training pipeline, automating data processing, and redesigning abstraction layers. You will also have the opportunity to lead a pioneering team with extensive experience in text-to-speech technology. We are looking for someone with expertise in PyTorch models, experience deploying and monitoring models in production, and a deep understanding of scaling machine learning services. If you are passionate about AI ethics, social impact, and fostering an inclusive team culture, this role is for you.

• Iteratively and incrementally improving the TTS service's ease of use, reliability, and performance.
• Building automated training and data processing services.
• Scaling the platform to load, process, and validate large amounts of text and speech data.
• Improving the performance and scalability of the machine learning training pipeline by integrating tools like DeepSpeed, PyTorch AMP, PyTorch JIT, and PyTorch Profiler.
• Researching techniques for better fine-tuning and quantizing models.
• Implementing processes to build confidence in the platform and services.
• Leading discussions on AI ethics and the social impact of the work.
• Fostering an inclusive team culture and working environment.
• Developing open-source reusable libraries and tooling.
• Developing complex PyTorch models.
• Deploying and monitoring models in production.
• Scaling machine learning services.
• Leading the development of a mature production machine-learning service.
• Deploying deep learning models with PyTorch.
• Building and deploying ML models for non-technical audiences.
• Paying great attention to detail.
• Creating modular, scalable, secure, and well-tested code.
• Understanding deep learning RND workflows, practices, and techniques.
• Using data structures and algorithms to process large amounts of data.
• Implementing automated model training and data processing.
• Implementing various tooling for speeding up training.
• Collaborating with cross-functional teams.
• Approaching problem-solving with an agile, iterative mentality.
• (Bonus) Building highly optimized layers with C++, CUDA, etc.
• (Bonus) Profiling and optimizing deep neural network performance.
• (Bonus) Scaling models past a billion parameters.

• Experience developing complex PyTorch models
• Experience deploying and monitoring models in production
• Deep understanding of scaling machine learning services
• Experience leading the development of a mature production machine-learning service
• Deployment of deep learning models with PyTorch in the past 2 years
• Building and deploying ML models for use by a non-technical audience
• Attention to detail
• Affinity for creating modular, scalable, secure, and well-tested code
• Deep understanding of deep learning RND workflows, practices, and techniques
• Background using data structures and algorithms to process large amounts of data
• Experience implementing automating model training and data processing
• Experience implementing various tooling for speeding up training
• Strong cross-team collaboration skills
• Agile, iterative mentality and approach to problem-solving
• (Bonus) Experience building highly optimized layers (with C++, CUDA, etc.)
• (Bonus) Experience profiling and optimizing deep neural network performance
• (Bonus) Experience scaling models past a billion parameters
• Must be a U.S. Citizen or Permanent Resident
• Must pass a pre-employment background check

• Competitive salary and stock options
• Full medical, dental, and vision insurance
• Matching 401(k) plan
• Generous vacation policy/paid time off
• Parental leave
• Learning & development stipend
• Home office stipend