Portfolio

---

About me

Responsibilities

• 4+ years of experiences as Software Engineer/Python developer
• Working on projects mainly dealing with AI/ML/Data Science/Analytics
• Perform EDA on client data, documenting the observation
• Building dashboard using Dash & plotly BI tools
• Deploying the model in the cloud, hosting web app for user interaction using Flask
• Creating demonstrative POCs for ML models
• Currently exploring knowledge over Deep learning and building end-to-end ML pipeline

IT Skills

• Programming - Python, R
• OS - Windows, Linux(Ubuntu, Centos )
• Database Knowledge of SQL along with joins, subqueries, views,
• NoSQL database like MongoDB, Cosmos DB
• Search engine Elasticsearch
• Data Analytics - EDA using R & Python (Pandas, Numpy, etc.)
• Data Visualization - Tableau, Power BI, Python(Matplotlib,seaborn,plotly,dash), ggplot,
• Machine Learning - Worked with various ML algorithm, prediction based projects, perform ensemble modelling
• NLP - Building chatbot using Dialogflow, Rasa NLU, chatterbot
• Web Development - Flask, Dash, Django, Streamlit
• Cloud Computing - Basic services & deployment of models on AWS, Azure stack: Azure function, azure devops
• Project methodology- Agile, Kanban
• Version control - Github, Bit bucket
• Tools/IDE - Jupyter lab/notebook, VS Code, Sublime, Postman

---

Projects

Music Time plugin for Sublime Text

Music Time is a plugin that discovers the most productive music to listen to as you code. It will recommend the songs using AI, and will provide control of Spotify player to the developer through the editor only

---

House Price Prediction Dashboard

Considering various features of house sale data, we build a predictive model using Gradient boosted regressor, which can estimate price of house.

---

FAQ Chatbot

A chatbot that will handle FAQs. We used Rasa core and Rasa NLU which are open source python libraries for creating conversational software that helps in making machine-learning based dialogue management and natural language understanding systems.Article

---

Face Recognition system

Facenet is also mostly deployed with MTCNN (for face detection and alignment) which is computationally expensive, so instead we prefer to use Haar Cascade for face detection which saves a lot of excessive computation and avoids incorrect recognition of distant persons. Instead of looping through user data, creating embedding every time, we had serialized them which has reduced the recognition time significantly.Article

---

Other works

• ImageBot

  A Telegram bot for searching image

• Fake profile generator

  Simple code for scraping website for fake profile data generation

• Rasa cheatsheet

  Quick guide for rasa chatbot framework file structure and command list

• Marathi MoodBot

  The project was a quick implementation of chatbot using the tensforflow embeddings. Though the language mentioned in config file is English but words are in Marathi langauge but written in English.

• Capture Crop Photo for Face recognition preprocessing

  A simple python code using in Face Recognition to capture photo using webcam, crop to image centered to face detected

• REST APIs Django

  Building REST API using Django-REST framework

• Elastic Stack Setup

  Steps for installation of Elastic stack on windows machine

• Elasticsearch: CRUD operations

  CRUD operations using Elasticsearch restful APIs

• Elasticsearch: Search-queries

  Search request APIs to search required data from indices of Elasticsearch

• Streamlit examples

  List of useful components of streamlit framework with code snippets

• Spark on Windows

  Step by step guide to install spark cluster on windows machine

---

Experiences

Organisation	Position held	From	To
Tata Consultancy Services	Backend Developer	March, 2022 onwards	-
S G Analytics Pvt. Ltd.	Data Engineer	May, 2021	February, 2022
Infogen-Labs Pvt. Ltd.	Sr. Software Engineer (AI domain)	October, 2018	April, 2021
Vuclip India Pvt. Ltd.	Intern (Data Analysis)	10 July, 2018	9 October, 2018
SGGS IET	Intern (CVPR)	22nd June, 2015	10th July, 2015

---

Education

Qualification	Institute	Percentage	Year of Passing
Post Graduate Diploma in Big Data Analytics	Know-IT, CDAC, Pune	72.88	2018
Bachelor of Engineering (I.T.)	SSBT’s COET, Jalgaon	62.90	2017

---