Computational Physicist with 10+ years of experience in programming and scientific computing. Proficient in system modeling and validation. Expert in force field development and molecular dynamics simulations (equilibrium and non-equilibrium). Solid understanding of object-oriented programming and software development. Highly motivated to learn new skills.

Hemanta Bhattarai Resume

1. dotFile : Configuration file for vim, tmux, bash shell.

2. OpenMD: OpenMD is open source molecular dynamics engine maintained by Gezelter lab. It is written in C++ and Python. Basic required files to run the simulation using openmd are force field file file_name.frc and system file system_name.omd. The forcefield file contains all the information about the force field parameters used during simulations and .omd file contain the information about the system and ensemble used.

3. GOBAD1.0: GOBAD1.0 is a package written in matlab and bash to analyze astronomical data to study the evolution of universe using spatial orientation of angular momentum of galaxies in a cluster.

4. Data science and ML projects: Some datascience and machine learning projects are found here.

   a. Audio Classification: Feature extraction from audio file to train the Artificial Neural Network and classification with the model accuracy of 0.88.

   b. Image Classificaiton: Image classification using Convulation Neural Network.

   c. Twitter Network: Use of Tweepy api to extract the followers and following for a use to generate a friendship network starting from the given user. The user id is uniquely transformed to different numbers to ensure the privacy of the user.

   d. Hot keywords in abstract: Extraction of the abstract from the papers published in aXrive using beautiful soup. Analysis of the extracted abstracts using natural language processing to predict top 10 keywords used by authors in abstract and see the research trend.

   e. Decision Tree: Implementation of decision tree classifier from scratch.

   f. Recommendation System: Movie recommendation system for movieLens data set using different similarity algorithms.

5. Lets learn something new: Contains random repositories to learn something new.

   a. CUDA: Simple implementation of CUDA to calculate correlation function and some vector arithmetics.

   b. d3js: Used d3js and basic java script to simulate the calculation of pi using montecarlo simualtion and uniform distribution of points in polar coordinates.

   c. Django: Diet track webpage developed using django. It has capabilty to store the information of user logged in using a username.

   d. Pi Using Bash: It is a bash script which uses most frequently used syntax in bash script to determine the value of pi. It can be used as all in one cheat sheet to review syntax used in bash scripting.

   e. Some advanced concepts of Python: Implementation of decorators, static method, class methods and magic methods.

   f. SQL: Basic implementation of SQL.

6. Research: Analysis scripts written in python and jupyter notebook for graduate research work.