Computer Engineer Customer Service
Resume:
Ming Wen
Computer Engineer
**** ****** **, *** ***, GOLETA, 93117-4416, USA
425-***-**** · ****************@*****.***
Date of birth
Skills
Deep Learning
Python skill
Fluent in C, C++, C#
AI algorithm
PyTorch
Fluent in Modelsim
Tensorflow
PCB Design
Software and Hardware
Safety Design
Customer Service
Matlab
Keil
Fluent in EAGLE
Languages
English
Chinese
Profile
Hardworking College Student seeking employment. Bringing forth a motivated attitude and a variety of powerful skills. Adept in various social media platforms and office technology programs. Available software languages include Python, Matlab, and C++. Mainly biased towards deep learning, proficient in using RNN, CNN, and LSTM algorithms. Good at image processing and have been exposed to TensorFlow/PyTorch framework. And use FPGA board design project, skilled in using Modelsim, Keil. Good at carefully diagnosing and evaluating problems and providing practical solutions.
Education
Bachelor, UCSB, Goleta
September 2017 — December 2021
Major: Computer Engineering
Internships
Hardware Engineer, CoolanyP, Seattle
August 2019 — October 2019
• Collaborated with back-end personnel to design structure and processes optimized for product parameters.
• Wrote code specifically for use in asynchronous architecture environments.
• Test and debug the hardware system
In school Project
Design of anti-collision warning device for eye surgery equipment(Hardware and Software Design)
September 2020 — June 2021
In this project, the main hardware used Jetson and Intel cameras. The biggest trouble I encountered was that it was difficult for the camera to monitor the position of the robotic arm and follow it. I added color recognition to the code and set the recognition area so that the center of the robotic arm is always at the center of the screen, and finally solved this problem. After this, I also used a Convolutional Neural Network (CNN) to simply train the robot arm and object distance, so that the camera can better capture the robot arm position to warn. Here is a link to the project's web page: https://sites.google.com/view/camera-arm-tracking/main-page Design an automatically loading and unloading trolley (Hardware Design)
September 2020 — June 2021
The main purpose of the project is to design a car that can load and unload goods automatically, mainly involving the design content of road image acquisition, calculation, drive control, and motor control. The road is paved with a black rubber line solid belt in obvious contrast with the light green bottom as the road boundary, and the black twisted line dotted belt is applied as the auxiliary guide inside the road. In our project, we used a Convolutional Neural Network (CNN) to capture road images to map raw pixels from the front-facing camera to the steering commands of the self-driving car. This allows our trolley to turn, walk straight, and unload along the sideline. The controller of the car is Raspberry Pi 3 Model B with a 1.2ghz 64-bit quad-core processor based on ARM Cortex-A53 architecture, and the software platform is Linux.
Auto-storyspeaker
September 2020 — March 2021
This project is to tell stories based on pictures. These stories are based on stories from romance novels. We first train a recurrent neural network (RNN) decoder on romance novels. Each paragraph in the novel is mapped to a skip-thinking vector. The RNN is then conditioned on the skip-thought vector and aims to generate the passages it has encoded. We also train visual-semantic embeddings between images and captions. In this model, captions and images are mapped into a common vector space. After training, we can embed new images and retrieve captions. Given these models, we use a function to balance the gap between the retrieved image captions and the passages in the novel. Finally, we can get a novel paragraph of suitable length based on the name of the function and the picture. Photo Style Transfer
December 2020 — March 2021
This is an ordinary style transfer project with fixed style and fixed content. The image is used as the training volume and the vgg16 model is used. After continuous training, the generated image will be consistent with the content of the content image, and will also be consistent with the style of the style image. In this process, the content consistency of the generated image and the content image must be guaranteed, so the content difference of the picture is measured by the feature map output by VGG16. In this process, a loss function is used to optimize the content consistency between the generated image and the content image according to the MeanSquaredError of the feature map output by the generated image and the content image. In the style image, I used the Gram matrix to optimize the style difference between the generated image and the style image based on the MeanSquaredError between the generated image and the Gram matrix of the feature map output by the function. Design a parking lot statistics device (Software Design) September 2020 — March 2021
This project includes the system has information collection module, used to collect the license plate information of vehicles around the entrance and exit channels of the parking lot; The gate control module is used to open the gate of the entrance and exit channel of the parking lot when the license plate information of the vehicle is received; The signal induction module is used to collect the signal generated when the vehicle passes through the entrance and exit channel of the parking lot; The parking space statistics module is used to count the total number of remaining parking spaces in the parking lot when receiving the signal generated by the vehicle passing through the entrance and exit channels of the parking lot. In this project, the main software language used is C++
Detecting tin droplet used for EUV source(support) September 2021 — April 2022
This is an in-progress project, a detecting system set up based on image processing used for EUV sources. The tin target motion and stability are displayed by image capturing and processing in real-time. The main purpose of this project is to perform modeling, data simulation, and data optimization based on the provided images to improve the stability of tin droplets. This project is still in progress, and I provide image processing assistance to this team
Contact this candidate