Udbhav Prasad
Welcome to my Portfolio!
đź“ś About Me
Hi, I'm Udbhav. I like working with data and AI.
I'm currently a 4th year Computer Science Student and Math Minor at Ryerson University.
I am always eager to learn and work!
For a more detail look into my skills check
out my Resume!
đź’» Work History
Cyber Security Analyst - Royal Bank of Canada
- Analyzed large amounts of data to produce reports to discover suspicious activity with Splunk}
- Updated intelligence knowledge base and threat profiles from discovered threats
Research Assistant - Ryerson University (St Michael's Hospital)
- Created visualizations \& neural networks(RNNs \& Transformers) for time series analysis and prediction on crypto market in Python
- Perform Biomedical Image data analysis to create effective image unmixing algorithms in MATLAB
Application Programmer - Ministry of Health
- Independently created 2 Proof-Of-Concept (POC) applications for distributed log streaming with scalability in focus; implemented on ministry servers and Azure Cloud Platform:
- Using Azure Event Hubs, Functions and Azure SQL
- Using Spark Streaming from Kafka & NiFi data flow management alongside HDFS
- Created a 6-node cluster where I installed/configured many (Big Data) software in cluster mode(Spark, Kafka, NiFi, Hadoop) intended for use by other developers and POC applications
- Created Python GUI application with openpyxl and tkinter to automate Excel reports' formatting for Ontario Disability Requirements which saved manual formatting of 100s of Excel files
- Performance tested two web servers with JMeter to determine maximum concurrent users resulting in clearance for Production use
- Wrote scripts in JavaScript and Python for GUI to change and backup web-pages in production server (Apache Server)
🔧 Skills
Languages: Python, SQL, Java, C++, Scala, HTML, CSS, MATLAB, R, Prolog, LaTeX
Big Data Frameworks: Spark, Splunk, Kafka, Hadoop, NiFi
Development Tools: Azure, Databricks, Docker, Linux, Git, CMake, JMeter — PyCharm, Clion, VSCode, Vim
Python Packages: PyTorch, PySpark, scikit‑learn, NumPy, Pandas, Flask, Matplotlib, Seaborn
Soft Skills: Communication, Self‑directed, Teamwork, Problem‑Solving, Adaptability, Time Management
🖥️ Projects
Transformer Implementations Package:
- Python Library for Transformer neural networks that are used for many machine learning tasks
- Published Package on PyPi
- Implemented/Trained multiple Transformer neural networks from scratch
- Trained and optimized the models implemented
- Models used in tasks such as Image Classification and Sequence‑to‑Sequence translation
- Models: Vision Transformer(ViT), Data efficient image Transformers(DeiT), GPT, BERT, Vanilla
GANs Implementations:
- Python Library for Generative Adversarial Networks that are used for Image Generation/Translation tasks
- Published Package on PyPi
- Implemented multiple GANs and Auto‑encoders from scratch
- Models trained and optimized producing realistic images
- Models Implemented: StyleGAN, Pix2PixHD, SRGAN, DCGAN, WGAN, SNGAN, Variational Autoencoder, etc.
Neural Style Transfer Web App
- Flask Web App containerized with Docker to combine the style from an image to another
- Through Web Interface you can customize training process and run transfer in back‑end