Hrishikesh Mandal

DSP engineer with experience developing robust signal processing algorithms for acoustic signal detection, classification, and analysis in noisy real-world environments.

Background spans audio feature extraction, multi-microphone signal analysis, and end-to-end validation of data-driven pipelines using MATLAB and Python

Graduate Research Assistant - Johns Hopkins School of Medicine

(2021-08 - 2026-08)

• Developed robust and scalable audio signal processing algorithms for detection and segmentation of marmoset monkey vocalizations in noisy, real-world acoustic environments, reducing processing time by 90%
• Implemented multi-channel microphone analysis methods for acoustic source discrimination in complex environments, using cross-channel signal comparison to support sound localization
• Built audio feature extraction and modeling pipelines for speech-like signal analysis, including spectral feature computation, fundamental frequency tracking, and application of PCA, UMAP, and DBSCAN for structure discovery and automated call classification, with an accuracy of 94%
• Led end-to-end experimental audio data acquisition and validation workflows, integrating synchronized audio/video recording systems and long-duration datasets to evaluate algorithm performance under real-world, non-ideal recording conditions

Course Assistant, Signals and Systems - Johns Hopkins University

(2022-08 - 2023-05)

• Led recitation sessions covering digital signal processing, linear systems, filtering, transforms, and frequency-domain analysis for undergraduate engineering students improving academic performance
• Guided students through quantitative signal analysis and system modeling problems, reinforcing core DSP concepts and practical problem-solving techniques
• Supported evaluation and technical review of signal processing assignments, providing feedback on analytical methods and implementation approaches

Graduate Research Assistant - Cleveland Veteran Affairs Medical Center

(2019-05 - 2021-08)

• Used cluster computing to optimize implanted electrode parameters to properly activate relevant muscles to produce walking motion within a biomechanical model
• Designed and validated algorithms for comparing and aligning multi-sensor motion capture signals, including estimation of center-of-mass trajectories from heterogeneous measurement systems
• Applied statistical learning methods including cross-validation, error analysis and learning curve evaluation to optimize model performance and robustness across experimental conditions

MS - Biomedical Engineering - Johns Hopkins University (2026-08)

MS - Biomedical Engineering - Case Western Reserve University (2021-08)

BS - Biomedical Engineering: Devices and Instrumentation - Case Western Reserve University (2021-08)