Senior Staff Audio Machine Learning Engineer

Boston Dynamics is seeking a Senior Staff Audio Machine Learning Engineer to support the Spot Sensing team in building the next generation of acoustic intelligence for industrial machinery. In this onsite role in Waltham, MA, you will help evolve Acoustic Change Detection (ACD) by transforming both beamformed and raw audio into signals that support machinery prognostics, and by unifying acoustic features with other modalities for prescriptive maintenance outcomes.

You will work across the full pipeline, from signal processing and model architecture through deployment practices for industrial customers, partnering with the broader Spot Apps organization to translate audio insights into machine health indicators and operator-ready recommendations.

Responsibilities

• Design, refine, and scale digital signal processing and machine learning pipelines for the Acoustic Change Detection (ACD) system.
• Lead improvement and modernization of the acoustic detection stack, including transitioning from unsupervised models (for example, BGMM) to state-of-the-art anomaly architectures and architecting end-to-end supervised multimodal pipelines for automated fault diagnosis.
• Partner with the broader Spot Apps organization to integrate acoustic insights with visual, thermal, and VLM/LLM data streams into unified machine health indicators, supporting both mechanical diagnoses and recommended corrective actions for plant operators.
• Benchmark, optimize, ship, monitor, and continuously improve the solution for industrial customers.

Requirements

• 7+ years of deep practical expertise in audio feature extraction, audio fingerprinting (e.g., Shazam, DMCA enforcement), and segmentation (e.g., DCASE), applying machine learning to audio data (beamformed and raw) for anomaly detection, preferably in industrial environments or similar problem domains.
• Advanced proficiency in Python and C++, with strong numerical and scientific stacks such as NumPy, SciPy, scikit-learn, and PyTorch.
• Expertise in both time domain and frequency domain data analysis.
• Expertise in multi-channel array processing, including beamforming (Delay-and-Sum, MVDR, null-steering), Direction of Arrival (DoA) estimation, and synthetic aperture techniques.
• Expertise in unsupervised and supervised classification architectures for anomaly detection under real-world conditions.

Technologies

• Python, C++, NumPy, SciPy, scikit-learn, PyTorch
• Shazam, DCASE, BGMM
• Delay-and-Sum, MVDR, null-steering, Direction of Arrival (DoA), synthetic aperture
• VLM, LLM, TypeScript

Benefits

• Medical
• Dental
• Vision
• 401(k)
• Paid time off
• Annual bonus structure

Bonus Points

• Hands-on experience in cross-modal fusion combining audio representations with computer vision, thermal imaging, or multimodal foundation models/VLMs.
• Background in rotating machinery health monitoring (bearings, gearboxes, pumps) and familiarity with mechanical vibration/acoustic failure standards.
• Familiarity with TypeScript and modern web services for integrating work into Orbit, the fleet management software.

Location: Waltham, MA (onsite)
Compensation: USD 180,000 - 220,000 per year
Minimum experience: 7 years