Principal ML Engineer

About the Company

Sanas is pioneering the future of human communication. Founded by a team of Stanford researchers and entrepreneurs, Sanas has developed the world's first real-time speech AI platform capable of accent translation, noise cancellation, and speech enhancement. The company is a fast-growing startup in Silicon Valley, backed by leading investors like Google Ventures and Insight Partners.

Responsibilities

• Architect robust, modular ML pipelines for model experimentation, feature extraction, and production inference
• Collaborate with data engineering to improve audio dataset quality, labeling pipelines, and feature engineering
• Mentor and collaborate with other ML engineers and research scientists to ensure best practices in model development, evaluation, and deployment
• Optimize models for latency, memory, and real-time performance on CPU/GPU/edge hardware
• Introduce frameworks for continual learning, model versioning, and A/B testing in production
• Stay current with advancements in Voice AI, Deep learning and multimodal model architectures

Requirements

• 10+ years of experience in Machine Learning Systems
• 3+ years in a technical leadership capacity
• Advanced proficiency in Python
• Proficiency in ML frameworks like PyTorch, TensorFlow, or JAX
• Strong understanding of Deep learning architectures (RNNs, LSTMs, CNNs, Transformers, CTC)
• Experience deploying ML models to production via ONNX, TensorRT, TorchScript, or custom inference stacks

Preferred Qualifications

• Familiarity with audio data challenges and time-series features
• Experience with Voice AI models such as ASR, TTS, and speaker verification
• Familiarity with real-time data processing frameworks like Kafka, Flink, Druid, or Pinot
• Experience with MLOps, feature engineering, and model training/inference workflows
• Background in DSP, audio augmentation, or working with noisy/multilingual datasets