Machine Manager

SRIKANTH MAGANTI

Chicago, IL adcg9x@r.postjobfree.com +1-708-***-****

https://linkedin.com/in/srikanth-maganti https://github.com/srikanthmaganti EDUCATION

University of Illinois at Chicago Chicago, Illinois Masters in Computer Science GPA: 3.8 Jan’19 - Dec’20 Computer Algorithms, Artificial Intelligence, Natural Language Processing, Machine Learning, Deep Learning for NLP Advanced Database Management Systems*, Data Science*, Information Retrieval* Chaitanya Bharathi Institute of Technology Hyderabad, India Bachelor of Technology in Computer Science and Engineering GPA: 8.3/10 Sep’13 - May’17 SKILLS

Programming Languages : Python, Java, C++, C, C#, .NET Big Data Skills : Hadoop, Map-Reduce, Spark, Scala Deep Learning Models : RNN, CNN, LSTM, GRU, Transformer Machine Learning Packages : Tensorflow, PyTorch, Scikit Database Technologies : SQL, Oracle, PL/SQL, MS Access Web Technologies : HTML, Javascript, CSS, Servlets, Apache Tomcat EXPERIENCE

Programmer Analyst, Cognizant (Hyderabad, India) May’17 – Dec’18

● Maintaining and querying Databases using MySQL, MS Access and programming experience

● Developed and maintained Smarts Application which is used to validate information of client’s customers

● Worked on “Smarts Application” using SQL, PL/SQL, Oracle Forms and generating reports

● Studying Data collection, debugging the issues in existing solutions and providing efficient solutions

● Supported the client for technical issues and process improvement by conducting several client meetings PROJECTS

Predicting time-to-failure of expensive industrial machines Jan’20 - Feb’20

● Worked on four kinds of time series data to classify whether a machine operates in Normal mode or Faulty mode

● Implemented a algorithm that would work on any machine by using comprehensive components of a single machine

● Used different deep learning models for multivariate time series forecasting and improved performance of breakout detection Detecting Financial Fraud using Machine Learning Jan’20 - Feb’20

● Implemented a machine learning model that predicts the probability of the first transaction of a new user is fraudulent

● Building a framework by end-to-end implementation of machine learning pipelines for feature engineering, and models

● Random Forest performed well to some extent but Balanced Random Forest outperforms in classifying fraud vs no fraud Scoring and Summarization of Movies/Products Reviews Oct’19 - Dec’19

● We used transfer learning over BERT and models like RNN, LSTM, GRU, Bidirectional LSTM to generate the sentiment of the products

● We worked on the well established Centroid based Extractive Summarization method to overall review of products

● The BERT-LSTM is shown to outperform existing algorithms for sentence classification Sentiment Analysis using different Deep Learning Models Sep’19 - Oct’19

● Sentiment Analysis on stanford movie review dataset by using simple RNN, LSTM, GRU, and CNN

● GloVe embeddings used as pre-trained word embeddings and achieved an accuracy of 81% with LSTM model Stock Prediction using Machine Learning Nov’19 - Dec’19

● Stock prediction using a class of powerful machine learning algorithms such as Ensemble Learning, Neural Networks

● Random Forest performed well on majority of the datasets whereas, XGBoost performed better by a small margin on unshuffled data Umbler’s Middle-Out Filter and Page Rank Implementation using Spark, Map-Reduce and Hadoop May’19 - Aug’19

● Evaluating number of unique three-word phrases followed by a set of nonfluency words in them

● Solved word-count and matrix multiplication problems over a huge volume of data using the map reduce programming model Natural Language Interfaces to DataBases Mar’19 - May’19

● Implementation of Algorithm for Natural Language questions into SQL queries by using Compositional Semantics and metadata

● Building sets of possible clauses to get the set of all possible SQL queries and finally producing an answer to the question Pacman using Artificial Intelligence Feb’19 - May’19

● Implementation of an AI-agent for Pac-Man and the ghosts, that can behave in an acceptable manner by using different AI techniques

● AI techniques like BFS, DFS, Greedy Search, UCS, A* Search, MDP, Value Iteration, and Q-learning

Contact this candidate