| | | 20,000+ Fresh Resumes Monthly | |
|
|
| |
Click here or scroll down to respond to this candidate Candidate's Name
Email: EMAIL AVAILABLE | Phone: PHONE NUMBER AVAILABLE | Github: BenyaminGithub EDUCATIONRutgers University(New Brunswick, NJ)
Bachelor of Science Degree in Computer ScienceSeptember Street Address - May 2024Relevant Courses:Artificial Intelligence Machine Learning Principles Brain Inspired Computing
Computational Robotics Operating Systems Software MethodologyData Management Systems Programming Data literacyAlgorithms Design & Analysis Minds, Machines, and Persons Discrete StructuresComputer Architecture Data Structures Linear Algebra
TECHNICAL SKILLSLanguages: Python, Java, C, C++, R, SQL, JavaScript, HTML, CSSFrameworks/Libraries: DuckDB, NumPy, Scikit-Learn, Matplotlib, PyTorch, JavaFX, Django, FlaskTools: Jupyter Notebook, Git, gdb, Android Studio, LaTeX, Google CollabWORK EXPERIENCEData Analytics ConsultantPrivate ContractorMarch 2024 - June 2024 Developed customized aggregation functions in C++ for DuckDB to address specific requirements of financial clients. Developed functions on Jupyter Notebook to create simplified SQL queries using arguments Utilized Scrooge-McDuck Extension https://github.com/pdet/Scrooge-McDuck Leveraged advanced programming techniques to optimize performance and ensure seamless integration with existing data analytics pipelines. Contributed to enhancing data processing capabilities, enabling clients to derive actionable insights from large datasets efficiently.PROJECTSRobot-Localization-Simulator Designed and implemented a robotic control and localization solution with four key components: Established obstacle-free environments and control sequences for robot navigation. Simulated execution with noise, generating ground truth poses and measurements for evaluation. Developed a Particle Filter-based localization algorithm with parameter experimentation. Conducted performance evaluation by comparing algorithmic estimates with ground truth poses. Utilized algorithms including a Kalman filter and a normal distribution (Gaussian distribution). Stored environments, controls, and measurements using jagged NumPy arrays. Visualized plots and simulations using Matplotlib. Tested different algorithms for likelihood of particles using Scikit-Learn, eventually implementing a custom solution.Classifying-Digits-and-FacesImplemented a Naive Bayes classifier for digit classification using PyTorch: Developed a custom dataset class for efficient data handling. Implemented dynamic training data adjustment to enhance classifier performance.Created a custom Convolutional Neural Network (CNN) for digit classification: Designed the CNN with 2 convolutional layers and 3 fully-connected linear layers.Employed a shallow Multi-layer Perceptron (MLP) for face and digit classification: Utilized one hidden layer in the MLP. Directly used raw pixels as features for both digit and face datasets. Described the MLP's suitability for handling complex non-linear mappings in digit and face datasets.Developed all PyTorch modules from scratch, leveraging PyTorch layers for implementation. |
