Job Description
📋 Description Ingest, clean, label multilingual audio data for ASR on-device use. Fine-tune and compress large models to fit iPhone constraints. Package language-specific weights for on-demand deployment. Handle loanwords and transliteration decisions in transcripts. Evaluate latency, accuracy, and robustness across languages. Document model cards and evaluation results for stakeholders. 🎯 Requirements Bachelor's degree in CS/DS/ML or related field. Strong data-engineering for large audio/text datasets. Experience fine-tuning/optimizing speech models (LoRA/QLoRA, quantization, distillation). ASR model development and multilingual evaluation with code-switching. Strong Python and SQL; experience with PyTorch, Hugging Face Transformers/PEFT, torchaudio, librosa. Production ML deployment and secure handling of data in regulated environments. 🎁 Benefits Generous and flexible time-off policy Flexible work schedules and telework options Career development and training opportunities Industry-leading benefits including health plans, FSA, commuter benefits 401K matching with company contribution Regular team events and social activities
