M.RAGHU VAMSI

Bengaluru · admlbq@r.postjobfree.com · 903*-***-***. https://github.com/Raghu29021996/raghuvamsi EDUCATION:

Master’s degree:

Vellore institute of technology. Chennai, India.

Mechatronics CGPA -8.11/10.

Bachelor’s degree:

August 2018-March 2020.

Avanthi institute of technology. Vizag, India.

Mechanical percentage -62.7/100. August 2013-May 2017. Skills:

Programming Language: Python, Matlab, Labview, Django, Html.

Technology: Azure Machine Learning Services, TensorFlow, Flask, Kera’s, SQL, Tableau, LabVIEW, Weka.

Scrapy Concepts: Text/Data Mining, Natural Language Processing (NLP), Deep neural networks, Computer vision.

EXPERIENCE:

Data science intern: Rubixe, Bangalore.

Inventory Management: August2020 - January 2021

Demand of spare parts of an automobile is always a challenge as most service centres spends significant amount time to get the parts from centre.

Developed a different model from a feature by considering different variables using time series ARIMA model.

Customer Transaction Prediction:

To Identify the customers who will make a transaction with the bank in future, irrespective of the amount of money transacted previously with the bank.

Applied domain knowledge, data wrangling techniques which transformed the dataset to suit customer segmentation by their purchasing patterns.

Which helped to achieve an accuracy of 91% using logistic regression. Automobile price prediction:

The model is to predict price of cars with the available independent variables. which makes the management to understand market with the independent variables.

Analysed the data and removed features which have less significance, built a model using Random Forest, linear regression.

Which helped me to get an accuracy of 98%.

Identification of handwritten numbers using CNN:

Different handwritten numbers and alphabets are considered for developing a data set and model was built using deep learning (Tensor flow).

With the test data and manual input, the model is verified. which gave me an accuracy of 94 %.

Walk or run:

Project which aims to detect whether the person is running or walking based on deep neural network and sensor data collected from iOS device.

Performed thorough exploratory analysis of the dataset by segmented of numerical information against the categorical features and correlation analysis. that bring more useful insights from the data which gave me accuracy of 98.97%. Projects:

Real Estate transaction price prediction model:

Performed thorough exploratory analysis, feature engineering, domain knowledge and developed Analytical Base Table (ABT).

Tested the fitted models with test set and checked R-Squared and Mean Absolute Error (MAE) metrics. Compared the hold-out scores and test scores and selected the winning model.

Real- time implementation of brake fault diagnosis using LABVIEW:

Created faults in brake system of car. To develop a model where data is acquired data using data acquisition system (DAQ). In form of vibrations.

Developed a model using random forest algorithm on Weka software. Which gave an accuracy of 95.4%.

Data-driven approach for retention management for a HR organization:

Conducted a thorough exploratory data analysis and performed bivariate segmentation analysis and visualization to understand the variables. And developed a analytical base table (ABT).

Checked AUC_ROC_SCORE for the fitted models by predicting with the test set.

Reviewed K-Means algorithm.

CERTIFICATIONS:

Data scientist certification from International Association for Business Analytics.

Participated in MECH-A-TON conducted by VIT Chennai collaboration with Valeo.

MATLAB and LABVIEW training course conducted by VIT.

HYPERMESH certification from Design tech Hyderabad.