Mehdi Karamollahi

I am a Senior Data Specialist at the University of Calgary, in Alberta, Canada. My expertise lies in identifying and visualizing patterns and trends in the data, interpreting the data to pinpoint issues, presenting solutions, and building predictive models and reporting dashboards. I have developed several advanced ML projects, leveraging large volumes of various types of institutional data with great results and accuracy. I also built multiple comprehensive reports and visualization dashboards for various business objectives. Currently, I am working on multiple other predictive models.

I have a Ph.D. in Computer Science from the University of Calgary. I successfully completed the program in 3 years and 2 months. I had the pleasure of working as a Graduate Research Assistant under the supervision of Professor Carey Williamson and Adjunct Professor Martin Arlitt (Principal Research Scientist at OpenText). During my Ph.D. thesis, I studied the pandemic effects on the University of Calgary's campus network. I worked on extensive volumes of data (more than half a petabyte) to study the longitudinal changes in network traffic patterns, the rise and performance implications of videoconferencing applications (with a deeper focus on Zoom), security implications on the campus network, and the changes in the Internet usage and behavioral patterns of different campus sub-communities.

Experiences

Education

Skills

• Python - extensive experience with Pandas, NumPy, SciKit-Learn, TenserFlow, Plotly Dash, Matplotlib
• Bash - extensive experience with scripting, automation, and pipeline development using HPC clusters and Linux-based systems
• Git - experience with GitHub for individual and team-based version control of projects
• Java and C++ - past experiences

• Vertica - extensive experience and highly familiar with this big data analytics platform
• MSSQL - extensive working and teaching experience
• PostgreSQL - highly familiar and good experience with this SQL flavor
• MySQL - limited experience in past projects
• MongoDB - limited experience in past projects

• SciKit-Learn - extensive experience developing various types of ML models
• TensorFlow - good experience developing various types of ML models
• PyTorch - limited experience
• Llama2 - good experience with running, tuning, and using the 13B models on a local system (currently working with 70B models in Cloud)

• Power BI - highly knowledged with extensive working and teaching experience
• Grafana - highly knowledged with extensive working experience
• Tableau - good working and teaching experience
• Plotly Dash - highly knowledged with extensive working experience
• Google Looker Studio - extensive working experience
• Matplotlib - extensive working experience
• Gnuplot - extensive working experience

• Azure - well familiar and good working and teaching experience with the ecosystem

Awards

• Simulation Modeling of Zoom Traffic on a Campus Network: A Case Study
  Mehdi Karamollahi, Carey Williamson, and Martin Arlitt.
  Performance Evaluation 162 (2023): 102382,



• A Retrospective on Campus Network Traffic Monitoring
  Martin Arlitt, Mehdi Karamollahi, and Carey Williamson.
  ACM SIGCOMM Computer Communication Review,



• Pandemic Effects on Campus Network Traffic
  Mehdi Karamollahi.
  Ph.D. Dissertation, University of Calgary's Digital Library.



• Packet-Level Analysis of Zoom Performance Anomalies
  Mehdi Karamollahi, Carey Williamson, and Martin Arlitt.
  Proceedings of ACM/SPEC International Conference on Performance Engineering (ICPE), Coimbra, Portugal, April 2023.



• Simulation Modeling of Zoom Network Traffic
  Mehdi Karamollahi and Carey Williamson.
  Proceedings of IEEE MASCOTS Conference, pp. 57-64, Nice, France, October 2022.



• Zoomiversity: A Case Study of Pandemic Effects on Post-Secondary Teaching and Learning
  Mehdi Karamollahi, Carey Williamson, and Martin Arlitt.
  Proceedings of Passive and Active Measurement (PAM) Conference, online, pp. 573-599, Springer, Cham, March 2022.



• Zoom Session Qualify: A Network-Level View
  Albert Choi, Mehdi Karamollahi, Carey Williamson, and Martin Arlitt.
  Proceedings of Passive and Active Measurement (PAM) Conference, online, pp. 555-572, Springer, Cham, March 2022, March 2022.



• Traffic Characterization of Instant Messaging Applications: A Campus-Level View
  Sina Keshvadi, Mehdi Karamollahi, and Carey Williamson.
  Proceedings of IEEE LCN, Sydney, Australia, pp. 365-372, November 2020.



• Measurement and Modeling of Tumblr Network Traffic
  Rachel Mclean, Mehdi Karamollahi, and Carey Williamson.
  Proceedings of IEEE MASCOTS Companion Workshop, Nice, France, November 2020.
  Appears in workshop proceedings, published as Springer Lecture Notes in Computer Science, Vol. 12527, pp. 237-253, January 2021.



• Characterization of IMAPS Email Traffic
  Mehdi Karamollahi and Carey Williamson.
  Proceedings of IEEE MASCOTS, Rennes, France, pp. 214-220, October 2019.

Other Interests