Research & Industrial R&D

Throughout my career, I have bridged the gap between academic theory and industrial application, managing projects ranging from real-time audio processing to large-scale machine learning architecture.

Route Stress Index for EV Fleets (AI Researcher)

(2025–2026) Volterra

• Objective: Developed a predictive model for Route Stress Index (RSI) to optimize electric vehicle fleet operations and resource allocation.
• Implementation: Architected an end-to-end supervised ML pipeline utilizing spatial-temporal telemetry and topographic data.
• Impact: Built scalable simulation frameworks to model operational workloads and delivered a production-ready prototype for early-phase deployment feasibility.

Wearable-Based Emotion Recognition (Post-Doctoral Research)

(2023–2024) OrbMedic & University of Ottawa

Focused on detecting affective states through wearable sensor fusion, prioritizing physiological and motion-based data to ensure non-intrusive monitoring.

• Sensor Fusion: Developed ML models integrating Electrodermal Activity (EDA), Electrocardiogram (ECG), Blood Volume Pulse (BVP), and inertial data (Accelerometer/Gyroscope).
• Innovation: Applied signal processing techniques to extract robust emotional indicators from noisy, real-world wearable data, avoiding the privacy concerns of computer vision.
• Applications: Targeted healthcare monitoring, workplace safety, and adaptive human-machine interfaces.

Incremental Open-Set Audio Recognition (PhD Thesis)

(2015–2022) University of Ottawa

My doctoral research addressed the “open-set” challenge—enabling systems to recognize known sounds while accurately identifying unknown audio events in dynamic environments.

• Contribution: Proposed novel algorithms for multi-class classification and incremental model adaptation for audio event detection and recognition.
• Dissemination: Resulted in three peer-reviewed publications validating performance across various levels of environmental “openness.”

Audio-Based Surveillance Systems (Research Contract)

(2016–2018) iWatchLife & University of Ottawa

• Development: Designed and evaluated real-time detection algorithms implemented in C/C++ for integration into smart home camera platforms.
• Innovation: Introduced a background–foreground audio separation technique to significantly increase detection robustness in real-world acoustic environments.
• Output: Published conference paper and released open-source MATLAB code for the research community.

LPC for Hearing Aid Technology (Research Contract)

(2015–2016) Starkey & University of Ottawa

Focused on low-latency speech coding architectures essential for modern hearing assistance devices.

• Technical Approach: Developed a Linear Predictive Coding (LPC) framework using backward adaptation and noise shaping to minimize bit rate and algorithmic delay.

Eukaryotic Gene Prediction (Master’s Research)

(2010-2011) The University of Texas at San Antonio

Conducted research in bioinformatics, applying statistical modeling and machine learning to genomic sequence analysis.

• Focus: Developing algorithms to identify coding regions (exons) within complex, intron-heavy eukaryotic structures.
• Skills Involved: Sequence analysis, biological signal processing, and computational genomics.