Home Resume
Resume
Cancel

Resume

Dinesh Naidu

Current Location: Johor Bahru, Malaysia
E-mail: dineshdna14@gmail.com
Linkedin: https://www.linkedin.com/in/dineshnaidu1014/
Availability: Upon Notice

Education

Universiti Teknologi Malaysia (UTM), Johor, Malaysia
Masters of Science (Data Science)
Current CGPA: 4.00
March 2022 - Present
Expected Graduation: October 2023

Universiti Teknologi Malaysia (UTM), Johor, Malaysia
Bachelor of Engineering (Mechanical)
2013 - 2017

Kolej Matrikulasi Kedah (KMK), Kedah, Malaysia
Pre-University
2012 - 2013

Working Experiences

Quality Engineer
February 2021 - May 2021
Flextronics Shah Alam Sdn. Bhd., Johor Bahru, Malaysia

  • Lead and resolve customer complaint investigations by using quality tools (5 Why & 8D Methodology)
  • Resolve quality issues by identifying problems, examining solution options, implementing action plans and providing resources
  • Utilise manufacturing quality tools such as PFMEA, Risk Management, Quality/Control plans
  • Continuous Improvement Activities (Internal/External/Customer Audit, CAPA)
  • Review, receive and analyze warranty returns (RMA & Refurbishing)

Quality Assurance Engineer
June 2018 - February 2021
Venture Pintarmas Sdn. Bhd., Johor Bahru, Malaysia

  • Enforce and maintain quality operation procedures relevant for the production site, e.g. process validation, non-conformance management
  • Customer complaints/issues investigation
  • Root cause identification and Corrective & Preventive Action
  • Internal Audit (Auditor & Auditee) of Quality Management System
  • Continuous Improvement Activities (Kaizen, Internal/External/Customer Audit, CAPA)

Process Engineering Intern
June 2016 - August 2016
Kyocera Telecom Equipment (M) Sdn. Bhd., Johor, Malaysia

  • Assist in day-to-day activities of jig and fixtures department to support production line
  • Ad-hoc tasks as required

Manufacturing Associate
June 2013 - August 2013
ASF Food & Beverage (M) Sdn Bhd

  • Involve in production line to meet daily output
  • Ad-hoc tasks as required

Professional Experiences

Technical Skills

  • Python (NumPy, Pandas, Scikit-Learn, RAPIDS, Matplotlib, Seaborn)
  • R
  • Jupyter Notebook
  • Machine Learning(Regression, Classification, Clustering), Data Mining
  • Microsoft Power BI, Tableau
  • Database (MySQL, MongoDB)
  • Microsoft Visual Studio Code
  • Exploratory Data Analysis (EDA), Data Visualization, Webscraping, Business Intelligence
  • Agile Project Management

Courses & Certifications

Business

  • Co-Founder of PNY Stickers
  • Small Custom Sticker Business operated on Lazada.com
  • In operation from June 2021 - December 2021

Key Projects

Crime Incident Mapping and Crime Rate Prediction Model using Machine Learning Tools

  • Master’s Degree Thesis Project
  • Extract, transform, and load crime incident, population, population density, inflation rates, unemployment rates, deprivation scores data in England
  • Development of a crime incident mapping dashboard using Microsoft Power BI
  • Development of crime rate prediction models using Random Forest Regression and XGBoost Regression for England from 2011 to 2022
  • The crime incident mapping dashboard provides an intuitive and interactive interface for exploring crime patterns and trends across regions and time periods
  • Leverages Microsoft Power BI’s capabilities to visualize crime data on maps and identify crime hotspots and areas of concern
  • Both Random Forest Regression and XGBoost Regression models achieved similar results in crime rate prediction, with an R-squared value of 0.85 and a Mean Absolute Error of approximately 2.8
  • Crime-related factors, living conditions, and socio-economic indicators as the most influential predictors of crime rate

EY - Open Science Data Challenge 2023 - Level 1

  • Global Semi-Finalist in the competition
  • Predict the presence or non-presence of rice crops at a given coordinate location (An Giang province in the Mekong Delta of Vietnam).
  • Utilize satellite data from Microsoft Planetary Computer via API. Primarily used MODIS and Sentinel-1-RTC satellite data
  • Use vegetation indices (VI) such as NDVI, EVI, DVI, SAVI, STVI, LAI and RVI for analyzing the greenness of vegetation
  • Additional statistical measures (mean, median, variance, standard deviation, range) computed for select VI
  • Apart from VI, latent heat flux (LE), total of evapotranspiration (ET), Fraction of Photosynthetically Active Radiation (FPAR) data were also used
  • Total number of features for a year is 239 features. Used data from 2020 and 2021, resulting in 478 features for training the ML model. Using multiple year data will help the model into taking account of different rice crop growth cycles in a year
  • Bounding box buffer used for each coordinate 0.0005
  • Experimented with different ML models: Logistic Regression, Support Vector Machine (SVM), Random Forest (RF) Classifier, Artificial Neural Networks (ANN)
  • Chose RF as the most accurate model, with ANN being close behind
  • Final model built using RF classifier with GridSearchCV. Managed to achieve 100% accuracy on the validation dataset

References

Shall be provided upon request

The resume as a PDF file is available for here