Sambeg Dhakal

Experience

• 

  Megasys Software Private Limited

  Feb 2018 - Jun 2019

  Database Developer/ FullStack Engineer

  • Built scalable and responsive web applications using ASP.NET MVC. Implemented modules for employee onboarding, leave requests, attendance tracking, and performance reviews. • Developed role-based access to enable features based on user roles. • Implemented ASP.NET Identity for user authentication and password management with multi-factor authentication (MFA) for enhanced security. • Designed, created, and managed databases and performed data analysis. • Initiated the use of PL/SQL that replaced the manual process and made use of Tableau for visualization, saving over 45% of work time. Show less

• 

  COTIVITI NEPAL (formerly Verscend Nepal)

  Jul 2019 - Mar 2022

  Software Engineer I

  ● Designed and implemented scalable RESTful APIs using ASP.NET Core Web API. Created interactive and detailed API documentation using Swashbuckle, enabling efficient testing, onboarding, and collaboration with developers. ● Integrated JWT authentication in .NET APIs to secure endpoints. and developed Role-Based Access Control (RBAC) policies with ASP.NET's built-in authorization features, restricting access based on user roles and claims. ● Investigated problems with SQL logic and ODI mappings created by developers, provided fixes, and ensured coding standards were followed. ● Tasked with locating and providing solutions for long-standing problems, which led to around a 50% boost in accuracy. ● Examined problems encountered throughout the transition from legacy mappings to Oracle Data Integrator (ODI) mappings and offered remedies to the team undergoing the migration. ● Guided, managed, and monitored newly recruited members of the team. ● Automated data analysis and quality control step using PL/SQL which reduced analysis time by 40%. ● Researched the value of the Oracle APEX tool and gave the team a demonstration for use in the future. Show less

• 

  Cotiviti

  Mar 2022 - Dec 2022

  Software Engineer II

  ● Designed and implemented scalable RESTful APIs using ASP.NET Core Web API. Created interactive and detailed API documentation using Swashbuckle, enabling efficient testing, onboarding, and collaboration with developers. ● Integrated JWT authentication in .NET APIs to secure endpoints. and developed Role-Based Access Control (RBAC) policies with ASP.NET's built-in authorization features, restricting access based on user roles and claims. ● Developed and maintained ETL (extract, transform, and load) using Actian Data Connect and SSIS for many retail audits, including Amazon, Walmart, Walgreens, Canadian Tire, and others. By streamlining current procedures and generating new mappings, data processing time was slashed by 30%.● Led as a primary engineer for the migration of several audits from Actian Data Connect mapping to SSIS mapping for its seamless integration with the SQL server. ● Designed, implemented, and modified PL/SQL that expedited data processing and cut down on query response times by 30%.● Automated processes such as data/file fixing, file merging, file breaking, file mapping, file loading, and various batch operations using technologies like "PowerShell," "Control-M," and ".Net" allowed for a 50% faster delivery time. ● Generated logic with T-SQL and SSIS to simplify data analysis and cut down on the time and effort required for manual analysis. ● Led a self-initiated research on the Outlook version for PST and discovered update is needed for automation which I presented to the team for permission, and after approval wrote a command line script, and used Control-M to execute and automate the process, reducing more than 8 hours of labor-intensive manual labor to a few minutes. ● Provided training to colleagues and new hires about "Audits", "PowerShell", "Control-M", "SSIS", ".Net", and Oracle Data Integrator mappings (ODI) Show less

• 

  Lamar University

  Jun 2024 - now

  ● Created three algorithms from scratch to categorize tweets based on their emotions o Two algorithms were employed to classify nine different emotions, each of which was further divided into positive, negative, and neutral categories. o created one more algorithm from scratch to categorize text as positive, negative, and neutral● Additionally, used a pre-trained rule-based sentiment analysis tool VADER to classify text. ● Neural network to further predict new tweets using the above results as a training and testing dataset. o Created 4 models using different conditions. o Work in progress to publish a paper.● Published a paper at an IEEE conference. o Paper name: “Sentiment Analysis and Emotion Detection of COVID-19 Geo-Tagged Twitter Data” o Initial sentiment analysis work using pre-trained tools and models like Corenlp Sentiment Score, Huggingface Sentiment Score, Flairnlp Sentiment Score, Labmt Happiness Score, and Vader Compound Score. o Also, calculated Covid-19 Related Anxious Tweet Ratio, Covid-19 Related Keyword Density.● Acquired experience with Python libraries such as NumPy, Pandas, TensorFlow, Keras, Pickle, Scikit- learn, NLTK, spaCy, wordninja, time, emoji, contractions, pathlib, and glob.● For research purposes used NVIDIA GPU with CUDA version ’12.2’. Carried out preprocessing in parallel. Show less

  • Graduate Research Assistant

    Jul 2023 - now

  • COAS summer camp counselor

    Jun 2024 - Jun 2024