Welcome to Prawar's Portfolio

Prawar Poudel, Data Scientist    IBM    Infrastructure Performance Team    11501 Burnet Road    Austin, Tx (78758)    Google Scholar Profile

Involvements

Professional Experiences

• International Business Machine (IBM)

Data Scientist, May 2021 - present

• Cloud infrastructure performance team
• Performance analysis of IBM cloud control plane
• Automation for error characterization, performance tracking
• Suggest optimization to developers
• Demonstrated performance improvements
• Awarded Outstanding Technical Achievement Award with the team
• Center for Cybersecurity Research and Education (CCRE), UAH

Intern

• Team Lead: Led research effort to suggest possible optimizations for archiving software. Conducted bi-weeklytechnical conferences
• Performance analysis: Performance evaluation of decompression and decryption algorithm for extraction ofarchives. Modification of software for implementing suggested optimizations. (Valgrind, gpof, C++, expect, bash)
• E & T Groups, Nepal

Software Developer

• GUI Development: Developed and added parts in the already existing GUI for MUJO(http://www.eandtnepal.com/) (Visual C++, MS Visual Studio)
• PreProcessing: Debugged CAD model to particle conversion module for MUJO, a simulation software. (C++)
• Particle Visualization: Developed a module to display particles with color-gradient based on their physicalproperties like temp, pressure etc. (C++/CUDA/OpenGL)
• Animation and Analysis: Developed function for analysis and display of CAD models of auto-parts on inclinedplane.(C++)
• E & T Groups, Nepal

Intern

• High Speed Data Transfer: Developed an application to transfer files among different branches of company.(JAVA)

Academic Experiences

• LACASA Lab, UAH

Research Student

• Research exploring physical properties of embedded flash memory in commercial microcontrollers.
• Research developing security primitives exploiting physical properties of flash memories.
• Developed firmware for interfacing commercial NAND flash memory chips.
• Developed firmware for interfacing commercial emerging non-volatile memories.
• Performance analysis of commercial emerging non-volatile memories.
• Department of Electrical and Computer Engineering, UAH

Graduate Teaching Assistant

Codes

Please find the codes written by me at the github link.

Education

• The University of Alabama in Huntsville

Ph.D. in Computer Engineering

Minor:Maths,Cybersecurity

• The University of Alabama in Huntsville

M.Sc. in Computer Engineering

Using nor flash memory in microcontrollers for generating true random numbers

• Tribhuvan University

Bachelors in Electronics and Communication Engineering

Skills

• Software Development: C, C++, Python, MATLAB, Expect(Tcl), MYSQL
• Hardware Development: System Verilog, Quartus, Platform Designer, Modelsim, Medici (Synopsys)
• Extensions and Libraries: Numpy, Pandas, Tensorflow, Keras, Sklearn
• Tools: Shell Scripting, GIT, MS Visual Studio, Wireshark, IDA, gdb, gprof, valgrind
• Platforms: x86, ARM, NIOS, MSP430, Flash Memory, Linux, Raspberry Pi, Windows, Google Coral

Courses

Courses Taken

• Computer Hardware and Architecture: Hardware Reliability, Advanced Computer Architecture, Hardware/Software Co-Design, VLSI Design, Advanced Embedded Systems
• Cybersecurity and Systems: Computer Network Security, Operating Systems, Virtualizations and Containers, Advanced Cybersecurity Engineering
• Data Science : Data Mining, Machine Learning, GPGPU Computing, Parallel Programming, Modeling and Analysis of Computer Systems

Courses Taught

Operating Systems, Introduction to Embedded Computer Systems, Digital Hardware Design Fundamentals, Introduction to Electronic Analysis and Design

Notable Projects

• ONFI NAND Flash Interface:

  • Developed research framework for exploring physical properties of flash memories.
  • ONFI NAND flash memory interfaced using ARM Cortex on DE1-SoC platform.
  • Code at: https://github.com/sickRanchez-c137/onfi_plus.
  • Initial documentation with interfacing in NIOS processor is found at https://sickranchez-c137.github.io/nios_NAND_interface/.

• SPI Memory Interface:

  • Developed a framework for interfacing emerging non-volatile memories using SPI protocol;
  • Raspberry-Pi was used as the host.
  • Codes can be found at https://github.com/sickRanchez-c137/Raspberry-Pi-SPI-Memory.

• Deep Inference in FPGA:

  • Developed a hardware accelerator for a deep neural network inference to be used with NIOS computer system.
  • The hardware and software code can be found at https://github.com/sickRanchez-c137/InferenceInFPGA.
  • A simple report can be found at this link.

• Breast Cancer Data Analysis:

  • Analysed different classification methods to perform classification of breast cancer data.
  • Dataset was obtained from Kaggle.
  • Code and documentation can be found at https://github.com/sickRanchez-c137/BreastCancerDataAnalysis.

Public Profiles

Coming soon............

Publications

• Prawar Poudel, Biswajit Ray, and Aleksandar Milenkovic. 2021. Microcontroller Fingerprinting Using Partially Erased NOR Flash Memory Cells. ACM Trans. Embed. Comput. Syst. 20, 3, Article 26 (May 2021), 23 pages. https://doi.org/10.1145/3448271
• P. Poudel and A. Milenković. 2020. Saving Time and Energy Using Partial Flash Memory Operations in Low-Power Microcontrollers. In 2020 21st International Symposium on Quality Electronic Design (ISQED), 183–189. DOI:https://doi.org/10.1109/ISQED48828.2020.9137034
• P. Poudel, B. Ray, and A. Milenkovic. 2020. Flashmark: Watermarking of NOR Flash Memories for Counterfeit Detection. In 2020 57th ACM/IEEE Design Automation Conference (DAC), 1–6. DOI:https://doi.org/10.1109/DAC18072.2020.9218521
• Prawar Poudel, Biswajit Ray, and Aleksandar Milenkovic. 2019. Microcontroller TRNGs Using Perturbed States of NOR Flash Memory Cells. IEEE Transactions on Compututers 68, 2 (February 2019), 307–313. DOI:https://doi.org/10.1109/TC.2018.2866459