Bland
Machine Learning Researcher, Audio
About this role
Bland is seeking an ML researcher to develop foundational audio technologies for enterprise AI phone agents, including speech-to-text, text-to-speech, and neural audio codecs. You'll move research from theory through large-scale training to production systems serving millions of daily calls, collaborating across research and engineering teams.
What you'll do
- Design and train large-scale text-to-speech models with expressive, controllable output and optimize for real-time inference
- Build robust speech-to-text systems handling accents, noise, telephony artifacts, and code switching
- Research and implement neural audio codecs achieving high compression with minimal perceptual loss
- Develop and scale distributed training pipelines for massive multilingual audio datasets
- Design rigorous ablation studies and experiments with both objective metrics and perceptual evaluations
- Curate audio datasets and implement data filtering and staged training strategies
What they're looking for
- Self-supervised learning and generative modeling
- Text-to-speech (TTS) system development and scaling
- Automatic speech recognition (ASR) and robustness techniques
- Neural audio codecs and audio compression
- Distributed GPU training and large-scale ML pipelines
- Experimental design and rigorous evaluation methodology
- Audio signal processing and multimodal modeling
- Python and deep learning frameworks
Benefits
- Remote or San Francisco office location
- Work on foundational voice AI research at scale
- Collaborate with well-funded company backed by leading Silicon Valley investors
- Opportunity to impact enterprise customer interactions through voice technology
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.
Bland
Bland builds AI voice agents for enterprise customers, handling millions of daily calls through custom deployments and proprietary audio and language technologies. The company is hiring Forward Deployed Engineers to work directly with clients on production solutions, and ML researchers to advance multimodal LLMs and audio technologies for real-time conversational AI.
View all jobs at BlandLikely interview questions
- Walk us through a TTS or STT system you've built or significantly contributed to—what were the key architectural decisions and how did you measure success?
- How would you approach improving ASR robustness to accents, noise, and telephony artifacts in a production system serving millions of calls daily?