Hello, I'm Manheng Wang
I love Software Engineering
I'm a Software Engineer with experience in Web Dev, Machine Learning and Database.
I love Data Analysis
I'm a Data Analyst with experience in ETL, Dashboard Development, and Operation Optimization
I love Teaching
I'm a Teaching Assistant of a Database class, feel free to ask me anything about query language and Database system ;)
Download CV"A great computer scientist and project partner somehow good at PingPong eat a lot of McDonalds" --Jiawei
About Me
I'm Manheng Wang at University of Michigan
I'm a Junior at the University of Michigan Studying Computer Science and Data Science. I'm currently looking for full time opportunities after my graduation in May 2024. Please reach out if you have any questions!
Email : manhengw@umich.edu
Linkedin : https://www.linkedin.com/in/manheng-wang-39a4aa197/
Education : University of Michigan College of LSA
Majors : Computer Science, Data Science
Phone : 810 543 5384
Citizenship: China
Currently Seeking : Full time SWE, Data Scientist, and Data Engineer positions (After April 2024)
Github : https://github.com/ManhengWang
SQL
C++
Python
R
C
HTML
CSS
Education
2019 - 2023
Bachelor's of Science in Engineering
University of Michigan College of Engineering
Computer Science Engineering and Data Science
Relivant Course work:
• Distributed Systems (EECS 491)| Grade: A-
• Operating Systems (EECS 482) | Grade: A
• Advanced Operating Systems (EECS 498) | Grade: A-
• Machine Learning (EECS 445) | Grade: A
• GPU arch and programming (EECS 471) | Grade: TBD
• Computer Organization (EECS 370) | Grade: A
• Data Structures and Algorithms (EECS 281) | Grade: A
• Foundations of Comp Sci (EECS 376) | Grade: A-
• Linear Algebra | Grade: A
• Calculus I - IV | Grade: A
• Discrete Statistics and analysis | Grade: A
2019 - 2023
Bachelor's of Business Administration
University of Michigan Ross School of Business
Relivant Course work:
• Financial Accounting (ACC 300)| Grade: A
• Managerial Accounting (ACC 301)| Grade: A-
• Business Law and Ethics (BL 300)| Grade: A-
• Operations Management (TO 313)| Grade: A
• Revenue Management (TO 470)| Grade: TBD
• Behavioral Economics (BE 300)| Grade: A
• Micro & Macro Econ (ECON 101 & 102)| Grade: A
Experience
2022 - 2022
AWS CloudFormation SWE Intern
• IAM Execution Role Generator: Proposed and designed a role permission generator for users to use within cloud formation. Resulted in a seamless process where users can generate roles on-the-fly with the appropriate permissions to use various third party extensions. Decreased the engagement time with software by 97% resulting in 15% higher adoption.
• Permission Validation: Designed a simulation which validated a customers permissions, which resulted in them being informed whether or not they could use a third party extension within CloudFormation.
• Contract Tests: Developed a mechanism to validate a publishers CloudFormation extension and the necessary permissions to use it. Resulted in improved customer clarity into what an extension was performing on their account.
2021 - Current
Computer Organization Instructional assistant
• Exam Creation: Led the design team for the EECS 370 midterm and final exam taken by over 800 students. Routinely assessed current learning objectives and integrated questions and feedback from other course staff.
• Discussion Leader: Led weekly discussions of 30+ Students. Created various online resources to reinforce project development knowledge (ARM Assembly Guides, C workshops).
2021 - 2021
IBM Storage Hardware Engineer
• Machine Learning Storage Vulnerability Recommender System: b Developed python-based collaborative filtering algorithm to compare various organizations in order to determine proper solutions to increase their storage resiliency, security, and operational efficiency within the scope of HPA/HPC software.
• Storage Resiliency Assessment: Lead the development of an automated flash storage resiliency assessment by utilizing past organizations resiliency vulnerabilities and developing an internal ranking alongside a recommendation of an IBM solution to eliminate the persisting problem
• RDMA InfiniBand (IB) Documentation: Developed material and taught several full-time solutions specialists about IB RDMA capabilities. Developed technical documentation for IBM storage solutions which incorporate IB
Projects
Paxos Based RSM
Designed a replicated key value service that is replicated using Paxos. Implementation included designing a fully functional paxos algorithm to gain consensus on a generic operation. Additionally, I designed the interface for clients and a Paxos RSM to communicate with these services. The paxos Key Value RSM communicates with paxos itself to propose client operations and gain a quorum on sed operation.
• Designed in Golang
Primary Backup Replication
Designed a key value service that is replicated over a primary and backup. Implementation included designing a fully functional view service that keeps the absolute state of the current primary and backup as well as the abstractions for a client to communicate with the view service. Implemented a replicated key value service using the view service.
• Designed in Golang
Virtual Memory Manager
Created a virtual memory manager which supports both file backed and swap backed pages. Processes address spaces are maintained by the pager and syscalls are administered through the pager interface. Processes may fork child processes with a non-empty virtual arena as well as create their own address spaces. A customizable memory size allows for pages to be evicted by the kernel using a clock queue to control eviction orders. Enable bits are also controlled by the kernel to dictate read and write permissions as well as the residency of a given physical page.
• Designed in C++ 17
Thread Manager
Created a thread manager to administer the creation, lifetime, and execution of various thread bodies on a multi-core system. Thread Library has support for conditional variables, mutexes, as well as generic threads. Inner process interrupts are also supported by the library, alongside various thread lifetime management operations, such as yield and join.
• Designed in C++ 17
Pipelined RISC-V CPU
Designed a pipelined RISC-V CPU compatable with data hazards in verilog HDL. RAW, WAR and WAW hazards are all avoided by this pipelined implementation. Branch predictions are static and assumed not taken to allow for optimizations to CPI and runtime.
• Designed in Verilog HDL
Support Vector Machine Multiclass Classifier
Developed a python algorithm to train a polynomial kernel Support Vector Machine to map sample text reviews to a label map consisting of positive and negative attributes while optimizing the hyperparameters. Calculated and developed multiple different models which maximized different metrics for a multiclass SVM using a one vs one approach in order to correctly classify reviews on a rating scale of 1 to 5.
• Designed in Python 3
Neural Network multiclass classifier
Developed a python algorithm to train a 7 layer Convolutional Neural Network with 2 pooling layers. The classifier was trained on a subset of 2000 images of 10 various dogbreeds and validated on 500 images with an accuracy of 89.5%. For training, I used greyscale images of the dogs as well as distorted/rotated images to generate various training material to ensure a high accuracy score.
• Designed in Python 3
Custom optional Library
Designed an 'optional' class in C++ to support optional data types [an extention of std::optional] Supporting generic optional type wrapping and unwrapping with overloaded operators, portability with many std::algorithm functions such as a optional safe std::accumulate, and Container safe and typesafe operations with optional generic data
• Designed in C++ 20
Misc Content
Here is a collection of some of the things I'm interested in/proud of!
I teach computer organization!
I made the EECS 370 Website!
I'm a Challenger League player!
I built my own electric skateboard!
Contact Me
Have any questions or want to connect?
Please reach out!
Call me!
269 930 4061
Location
United States
maximosnolan@gmail.com
Website
www.maximosnolan.com
shorturl.at/pNS57