Respected Sir / Madam,
I take this opportunity to introduce myself, Namit Agarwal. I currently read as a 4th year student at Amity University Kolkata. I have worked for FinTech companies like OneWalletHub and Renderbit Technologies.
I wish to join an organization, which operates in a professional, dynamic and vibrant environment. Moreover, I am sure that there would be extremely challenging and growth oriented career opportunities existing in your organization.
On my part I can assure you that with my qualification and experience accompanied with the knowledge and skills, I would be able to make a valuable contribution to the organization, especially in the areas of Machine Learning and Data Science.
Please find enclosed herein my Curriculum Vitae for your kind perusal. Thanking You,
Encl: Curriculum Vitae
Mobile No: +91-990*******
To work in a challenging and a dynamic environment and be a catalyst to the change in management. I would welcome an opportunity to consolidate and expand this quest for knowledge and in process develop methods and solutions resulting in improved, efficient and effective work process. ACADEMICS
Examination Institute Year Percentage
BTECH Amity University 2020 7.43 cgpa
XII (CBSE) Birla High School 2015 90%
X (ICSE) M.C.K.V 2013 92.8%
• Programming Languages – Python, R,Java
• Database Language – MySQL
• Web Scrapping – BeautifulSoup ( Python)
• Version Control - Git
• Data Analytics- Python, R, Tableau.
• Web Dashboard – R Shiny
• Machine Learning - Python, Numpy, Panda, Seaborn, scikit-learn
• Deep Learning – Tensorflow, Keras
• Cloud – Google Cloud Platform
• Applied AI course – Learnt end to end about Machine Learning starting from EDA, dimension reduction and visualization, applied mathematics, NLP, supervised and unsupervised learning, feature engineering, productionization and deployment of ML models. It also gave me a complete mathematical background of Deep Learning and Neural Networks. *All the case studies I mention here were completed with the help of this institution. (Certificate pending)
• Data Science with R – This was a course on Data Camp which helped me revisit the above- mentioned concepts with R.
• OneWalletHub -
Position – Data Analyst Intern
OneWalletHub is a FinTech startup that offers a unified digital payments platform for retail using cutting-edge technologies such as Block chain for enhanced security and AI & ML for deeper insights into merchant & customer payment behavior. My role was to extract various insights from the transaction dataset. I've created wire-frame models and latter went on to write the algorithms for the finalized plots using R. I've also created the dashboards for both bank-side and merchant-side using R Shiny. Key Takeaway - I got an end-to-end experience in handling data, preprocessing it, extracting insights and creating the UI of the dashboard.
• Renderbit Technologies –
Position- Data Science Intern
I was honored to work one-to-one with Mr. Hirak Mukherjee sir, ex-Vice President of IBM, under the guidance of whom I got myself involved in option trading. Starting from the very basics I worked towards putting a strategy into algorithmic form. The extent of the project cannot be elaborated, as it has not been completed yet.
Key Takeaway - I got myself acquainted with the stock market where I've spent almost 20 days observing the movements of nifty and India Vix and analyzing what bracketing could be done to train our model.
I also learnt how to scrape data from web (in python using BeautifulSoup) by scrapping 2 sites- NSE INDIA and edelweiss.
• Donor Choose Dataset-
I’d worked on this dataset in my entire journey of learning Machine Learning. I’ve tried to build various Classification and Regression models on this dataset. The models include- k-NN, Naive Bayes, Linear and Logistic Regression, Support Vector Machines, Decision Tree, Random Forest(XGBOOST).
I’ve also tried Clustering the dataset where I tried to use the following clustering methods - kmeans, Agglomerative and DBSCAN. I’ve tried reducing the dimensions of the datasets using following methods - PCA, t-SNE, SelectBest, Truncated SVD. I’ve also used WordCloud to study text features.
I’ve preprocessed the dataset, added various features, prepared categorical, numerical and text features separately by vectorizing it, hyper parameter tuned all the models to its very extent using for loops, GridSearch and RandomSearch, and used confusion metric and heatmaps to study the results.
• Quora Question Similarity Problem-
The dataset was taken from Kaggle and studied. Constraints were found. Different EDA techniques were used to study different features. Text preprocessing was done and various features that could be extracted from the text were calculated and added. Text vectorization was done and the best of Bow, tfidf and word2vec was chosen. ML models were built on the prepared dataset and results were studied using confusion metric. Word Clouds were plotted.
• Personalized Cancer Diagnosis-
The dataset was taken from Kaggle and studied. Constraints were found and performance metric was chosen. The categorical features were one hot encoded and random encoded. The text feature were vectorized and various ML models -LR, NB, RF, XGBOOST, Stacking Classifier were built on the prepared data. Summary was stated with the help of PrettyTable.
• Facebook Friend Recommendation -
Dataset was taken form Kaggle. This was a link prediction problem were I’d to recommend friends to users. The challenge here was that only the user source and destination node ids were mentioned and links had to found. Various features were found using graph and network and the problem was posed as a Binary Classification problem. Various ML models were built on the prepared dataset.
• Stack Overflow-Tag Prediction -
The dataset was taken from Kaggle and the task was to predict as many tags as possible based on the given question. This case study helped me understand and learn how to work with Multi-label classification. I tried mapping the multi label classification to binary classification and then build models.
• Amazon Fashion Discovery Engine -
The task was to build a Content based Recommender System. Various features were formulated. Text based product similarities were computed on different vectorizations and compared later.
• Netflix Movie Recommendation -
The dataset was taken from Kaggle. Here I’d to build a Collaborative Recommender System. Similarity matrix was created for the ratings both user-user and movie-movie. XGBoost and Surprise was together used to build the system.
• Human Activity Recognition-
The UCI_HAR dataset was taken from Kaggle and LSTM and CNN models were tried to build. I also implemented the Divide and Conquer method on this dataset( based on a research paper) and bette results were obtained. All the hyper parameters were well tuned. ACCOMPLISHMENTS
• I have won state level championships in skating for 4 years straight in all three forms.
• I have been the winner of Inter-department Table tennis competition in my college for last two years.
• I was the finance head of my college fest AMIPHORIA.
• I’ve been leading my department’s team as Captain of Table Tennis for the last 3 years in our college sport’s fest Sangathan.
Father’s name : Sanjay Kumar Agarwal.
Date of birth : Sept12, 1997
Current address : 268/2 G.T. Road Liluah Howrah -711101 Language known : English, Hindi, Bengali & German
Hobbies : Playing Table Tennis & Cricket,Travelling,Listening Songs and solving Sudoku
LinkedIn : https://www.linkedin.com/in/namit-agarwal-012846145/ Place: Kolkata NAMIT AGARWAL