Signal ML and Acoustic Analytics
Signal processing, acoustic signal processing, vibration diagnostics, spectral analysis, envelope analysis, feature extraction, time-series modeling, anomaly detection, and degradation modeling.
Senior AI/ML Scientist - Trondheim, Norway
Muhammad Zohaib Sarwar, PhD
Signal Processing | Acoustic Analytics | Predictive Maintenance | Uncertainty-Aware Industrial AI
I build practical machine learning and signal analytics workflows for industrial assets across energy, rail, manufacturing, and infrastructure. My work turns high-frequency operational sensor data into validated diagnostics, calibrated model insight, and decision support for real-world maintenance and monitoring.
Open to
Senior Signal Processing / Acoustic ML Engineer
Preferred track — signal ML, acoustic analytics, vibration diagnostics, and physics-informed condition monitoring for industrial assets.
Senior Computer Vision Engineer / Scientist
Open to vision-based inspection, defect detection, and sensor fusion roles where industrial AI and evaluation rigor are central.
Principal AI / ML Scientist
Open to principal-level roles where uncertainty-aware decision support, model validation, and deployment-grade analytics are the core challenge.
Consulting and Freelance
Available for project-based work in predictive maintenance, signal analytics, operational AI, and model validation across energy, manufacturing, rail, and infrastructure.
Profile
I am a senior AI and machine learning scientist with a PhD in Engineering from NTNU and experience developing deployment-focused analytics for industrial assets. I work across signal processing, acoustic analytics, predictive maintenance, anomaly detection, uncertainty quantification, model validation, and production-oriented dashboards.
My applied work spans high-frequency operational sensor data, vibration-based condition monitoring, acoustic signal analysis, control system signals, IoT workflows, and structural health monitoring — across sectors including rail, energy, manufacturing, and infrastructure.
Expertise
Industrial AI built around signals, validation, and action.
Uncertainty-Aware AI
Probabilistic deep learning, uncertainty quantification, confidence scoring, calibration, robustness testing, out-of-distribution behavior, and safe automation thresholds.
Industrial Deployment
Predictive maintenance, condition-based decision support, operational analytics, industrial IoT, model validation, stakeholder reporting, and production-ready workflows.
Evaluation Rigor
Experiment design, baseline comparison, performance tracking, reproducible pipelines, version-controlled experiments, and peer-reviewed scientific code.
Cloud and Engineering Tools
Python, SQL, C/C++, MATLAB, Scikit-learn, TensorFlow, PyTorch, NumPy, Pandas, Azure, AWS, Docker, Git, Power BI, Apache Spark, COMSOL, and ABAQUS.
Collaboration
Cross-functional engineering, industrial partner coordination, technical documentation, stakeholder communication, and mentoring students and junior researchers.
Experience
From research-grade models to industrial monitoring workflows.
May 2025 - Present
Senior Data Scientist
Develop and validate data science workflows for real-time footfall analytics, audience segmentation, and campaign measurement across IoT and Wi-Fi sensor networks deployed in 30+ countries. Work spans sensor data calibration, probabilistic modeling, anomaly detection, and time-series forecasting on passive sensor streams to deliver reliable audience metrics for media and retail clients.
Sep 2023 - Present
Industrial Postdoctoral Fellow
Develop AI and signal processing workflows for railway condition monitoring using high-frequency operational data from onboard train systems. Architect scalable data pipelines across multiple train units, build Python applications for noisy real-world signals, and create Azure-hosted dashboards for diagnostics, degradation insight, and maintenance support.
2025 - Present
Contractor, Vibration Analysis and Machine Learning
Build signal processing and machine learning workflows for machinery diagnostics, vibration-based condition monitoring, industrial asset health assessment, and customer-facing reporting outputs.
2025 - Present
Freelance Data Scientist / ML Consultant
Design preprocessing, feature engineering, model comparison, validation, forecasting, anomaly detection, and decision support workflows for real-world industrial and commercial datasets.
Jan 2024 - Mar 2024
Engineering Intern
Developed Python tools to extract live control system signals from ABB AC 800PEC systems, configured data interfaces, validated signal quality, and documented repeatable workflows for monitoring and analytics.
Jan 2024 - Mar 2024
Engineering Intern
Supported commissioning, subsystem validation, troubleshooting, software and firmware configuration checks, and practical validation for asset-level monitoring requirements.
Sep 2022 - Jul 2023
Industrial Researcher
Built cloud-connected monitoring workflows integrating IoT sensor systems, analytics pipelines, and reporting tools for industrial process monitoring and software-supported engineering decisions.
Sep 2019 - Aug 2022
Doctoral Researcher
Developed probabilistic ML models for structural condition assessment, including preprocessing, feature extraction, anomaly-sensitive inference, uncertainty quantification, validation, and robustness evaluation.
Aug 2017 - Apr 2019
Research Assistant
Designed multi-sensor acquisition systems for long-term monitoring, embedded sensing, low-power wireless systems, signal processing, and field-oriented data collection workflows.
Selected Projects
Outcome-focused work across industrial assets, rail systems, and infrastructure.
2025 - Present
Industrial Condition Monitoring Workflows
Physics-informed signal processing and machine learning workflows combining sensing, feature engineering, anomaly detection, and structured reporting for machinery health applications across industrial sectors.
2024 - Present
Degradation Modeling from Continuous Operational Signals
Feature engineering and analytical pipelines linking continuous measurement signals to degradation behavior and maintenance insight — applied to rotating machinery and rail wheel wear as a case domain.
2023 - Present
Operational Diagnostics and Monitoring Applications
Python and Azure-based software and analytics workflows using high-frequency operational engineering data to support diagnostics, KPI tracking, and decision support for complex industrial systems.
2022 - 2023
Cloud-Connected Industrial Monitoring
IoT-based monitoring and analytics workflows integrating sensor systems, cloud pipelines, and reporting tools for industrial process improvement and engineering decision support.
2018 - 2019
UAV-Based Infrastructure Inspection
Drone-based inspection workflows and deep-learning visual defect detection methods for infrastructure condition assessment and structural health monitoring.
Consulting & Collaboration
Reach out for industrial AI, automation, and signal analytics opportunities.
I am open to consulting, freelance projects, technical collaboration, and senior AI/ML opportunities where reliable automation and engineering-grade model validation are central to the work.
How I can help
I help teams move from raw operational data to practical decision support: data extraction, signal processing, automation workflows, predictive models, uncertainty-aware evaluation, dashboards, and technical reporting.
Automation & Operational Analytics
Python workflows for data extraction, cleaning, signal processing, repeatable reporting, dashboards, and decision-support pipelines.
Predictive Maintenance
Condition monitoring, vibration and acoustic diagnostics, degradation modeling, anomaly detection, and maintenance insight for industrial assets.
Model Validation & Reliability
Baseline comparisons, uncertainty scoring, calibration checks, robustness testing, performance tracking, and safe automation thresholds.
Core Responsibilities I Can Own
- Signal and sensor data pipelines
- Industrial AI proof-of-concepts
- Monitoring dashboards and reports
- Technical documentation and stakeholder delivery
Selected Publications
Peer-reviewed work behind the applied practice.
Selected publications are listed here for relevance. See Google Scholar for the full record.
View full publication recordProbabilistic autoencoder-based bridge damage assessment using train-induced responses
Sarwar M.Z., Cantero D. Mechanical Systems and Signal Processing, 2024.
Vehicle-assisted bridge damage assessment using probabilistic deep learning
Sarwar M.Z., Cantero D. Measurement, 2023.
Deep autoencoder architecture for bridge damage assessment using responses from multiple vehicles
Sarwar M.Z., Cantero D. Engineering Structures, 2021.
Clustered Federated Learning for Population-Based Structural Health Monitoring
Cheema M.A., Sarwar M.Z., Rossi S., Cantero D. IEEE Internet of Things Journal, 2025.
Computationally-efficient structural health monitoring using graph signal processing
Cheema M.A., Sarwar M.Z., Rossi S., Cantero D. IEEE Sensors Journal, 2024.
Education
Engineering foundation with applied ML depth.
2019 - 2023
PhD in Engineering
Norwegian University of Science and Technology, Trondheim, Norway
Thesis: Vehicle-assisted bridge damage assessment using probabilistic deep learning.
2017 - 2019
MSc in Engineering
Chung-Ang University, Seoul, South Korea
Thesis: Event-driven structural displacement estimation using a multi-metric ultra-low-power wireless sensing system.
2013 - 2017
BSc in Electronic Engineering
Bahria University, Islamabad, Pakistan
Thesis: FPGA based sea range finder.
Languages
EnglishProfessional working proficiency
NorwegianA2/B1 — actively improving
UrduNative
KoreanBeginner
Contact
Industrial AI, predictive maintenance, and signal analytics.
Focused on senior AI/ML roles and collaborations where signal data, uncertainty-aware evaluation, and reliable industrial decision support matter.