AI Scientist – Natural Language, Health Records
Artisight
IoTPyTorchTensorflow
Job Level
Mid-LevelSenior
Tech Stack
PyTorchTensorflow
About the role
- Train, fine-tune, and optimize speech and audio foundation models, including large-scale and domain-adapted models.
- Contribute to the architecture of Artisight’s AI platform, ensuring seamless integration of audio intelligence with other AI modalities.
- Collaborate closely with AI scientists, engineers, and product teams to translate research ideas into production-ready solutions.
- Deploy Audio AI technologies into production healthcare environments and ensure low-latency performance where required.
- Stay up to date with advances in speech processing, generative audio, and multimodal AI and incorporate insights into the applied research pipeline.
- Share research outcomes through internal discussions, technical reports, and potentially external publications.
Requirements
- M.S. or Ph.D. in computer science, electrical engineering, applied AI, machine learning, or related discipline.
- Demonstrated expertise in speech and audio AI research, evidenced by open-source contributions or peer-reviewed publications (e.g., ICASSP, INTERSPEECH, NeurIPS, ICML).
- Hands-on experience with one or more of: Automatic Speech Recognition (ASR), Text-to-Speech (TTS) and voice synthesis, Audio-to-Audio / Speech-to-Speech Generation, Audio classification and event detection, Voice Activity Detection (VAD).
- Proficiency in deep learning techniques such as transformers, diffusion models, self-supervised learning, and sequence-to-sequence architectures.
- Strong coding and experimentation skills with frameworks such as PyTorch or TensorFlow.
- Experience with large-scale training and deployment tools (NVIDIA Triton, ONNX, or similar).
- A collaborative mindset and ability to communicate research findings clearly to both technical and non-technical audiences.
- Nice to haves: Experience with multimodal learning (audio + vision + text); Familiarity with federated learning and privacy-preserving AI approaches; Experience deploying real-time or low-latency audio AI models; Contributions to open-source audio AI projects (e.g., ESPnet, Kaldi, Hugging Face Transformers, Fairseq, Whisper).
