Hi, I'm Darshee.
I'm a Computer Engineer residing in the New York
Metropolitan Area ., seeking new opportunities
to contribute my skills and knowledge.
Introduction
Overview.
As an experienced computer science professional with a Master's Degree in Computer Science from Rutgers University, I bring a strong foundation in Software Development, Data Engineering, Cloud Computing and Machine Learning. I am proficient in Python, TypeScript, JavaScript, Java, C# and Golang. Throughout my journey, I've worked with various frameworks like Flask, Django, Node.js, Spring Boot and Express.js. I'm a quick learner, I thrive on collaboration and enjoy creating efficient, scalable, and user-friendly solutions to tackle real-world problems. I'm eager to work together and contribute my skills to bring your ideas to life. Feel free to reach out and let's explore how we can create something great together!
LANGUAGES
Python
JavaScript
TypeScript
Java
SQL
Golang
C#
C++
HTML
Shell
SolidityFRONTEND
CSS
React
Vue.js
Angular
JavaFXBACKEND
Flask
Django
Spring
Node.js
Next.js
Express.js
Crow
Firebase
Auth.jsDATABASES
PostgreSQL
MySQL
MongoDB
CouchDB
Neo4j
Elastic Search
Firestore
SnowflakeINFRASTRUCTURE
AWS Cloud Services
Azure Cloud Services
Google Cloud Platform
Docker
Kubernetes
GitHub
GitLab
Linux
EthereumTOOLS
Visual Studio Code
IntelliJ IDEA
GraphQL
Airflow
Jira
ChatGPT
GitMachine Learning
Pandas
Numpy
Matplotlib
Tensorflow
PyTorch
Sci-kit Learn
KerasGenerative AI
OpenAI
LLama2
Mistral
Langchain
ChromaDB
PineconeWhat I have done so far
Work Experience.
Data Scientist
Rutgers Institute for Health
- Spearheaded research on predictive cognitive disorder detection using keystroke and accelerometer data collected from smartphones, achieving a RMSE score of 0.18 using recurrent neural networks.
- Achieved 86% accuracy in predictive time series ECG data analysis through advanced machine learning algorithms.
- Cleaned and preprocessed ECG data using adaptive filters to remove 93% of motion artefacts.
- Streamlined data wrangling tasks and automated repetitive processes by making a data ingestion pipeline.
- Published a symposium and presented my research at the International Neuropsychological Society 2023.
Software and Data Science Intern
Chera Health
- Built an Android and iOS personalized food diary mobile app aiding 500 users with eating disorders in accessing tailored diet plans and specialized licensed therapists.
- Designed in-app forms to handle user feedback, gather daily health checks and raise support tickets using Java and C#.
- Leveraged users’ diet and mood data to optimize patient care plans and identify mental health triggers with an accuracy rate of 78% by deploying diurnal analysis using machine learning algorithms like Support Vector Machines.
Software Engineer
ZS Associates
- Adept at utilizing Jira Scrum Agile development for daily workflows.
- Utilized cloud microservices with Flask & AWS Lambda to streamline solutions, reducing response times by 50%.
- Dockerized applications firm-wide, enhancing deployment speed, scalability & quicker migrations.
- Maintained responsive 100 TB large big data systems and ETL pipelines for data warehousing.
- Led the Metadata Marketplace project by building a pipeline orchestrated using Airflow that generated metadata from structured data from over 20 data catalogs, enriched it, and published it to a Neo4j knowledge graph.
- Reduced overhead by 45% by leveraging GraphQL APIs for search & data discovery purposes.
- Implemented RESTful APIs using Python and Flask in conjunction with Postgres and MongoDB.
- Employed continuous integration and deployment (CI/CD) pipelines and Machine Learning to automate build and deployment processes, reducing release cycle times by 20%.
Cyber Security Software Engineer
Colgate Palmolive
- Developed and maintained a software scan portal using Django to validate software installation by deploying VirusTotal APIs which resulted in a 30% reduction in security tickets.
- Monitored safety incidents using vulnerability scanners like Qualys and OWASP ZAP and mitigated security issues using the a SOAR (Security Orchestration, Automation, and Response) platform like Siemplify.
Full Stack Developer
K.J. Somaiya College of Engineering
- Programmed a secure website for online tests with React for the frontend and Crow(C++) as the backend.
- Optimized database querying times, reducing latency by 25% to give a smoother user experience by implementing caching using Redis, partitioning data tables with AWS DynamoDB and indexing.
My work
Projects.
I have almost three years of experience; I worked as a software engineer, then as a data engineer, and I am currently working as a data scientist. I have a holistic understanding of applications and the data that influences them. I am really interested in working on real-world projects that have a meaningful impact on people's lives. Due to this background, I have a collection of projects that effectively demonstrate my skills and experience, showcasing real-world examples of my work. Some projects have demonstration links as well as repositories containing the source code. These projects serve as a testament to my proficiency in tackling different challenges, adeptness with diverse technologies, and ability to efficiently manage projects.
BuyMe- Auctioning Web Application
Engineered a highly scalable e-commerce full stack application, achieving a peak concurrent user count of 10,000 during peak hours, facilitating real-time auctions, bidding, and selling. Incorporated AWS Elastic Load Balancer (ELB) to distribute incoming application traffic across multiple targets to ensure high speeds and availability and 98% uptime. Provided real-time sales analytics through the dashboard functionality made with Streamlit.
ApplyByAI- Recruitment Platform
Architected a tool that screens resumes and takes preliminary HR interviews via a chatbot for applicant screening using IBM Watson Assistant to design over 50 Intents for interview questions. Proposed novel Employee Recommendation System based on Big5 traits, skill-matching & large scale hybrid filtering achieving recruitment match accuracy of 78% for top 5 recommendations.
AR-NAV- Indoor Navigation Application
Developed an indoor navigation app that leverages AR to point out the path by augmenting virtual "breadcrumbs". Used computer vision to object tag 75% of the surroundings and create indoor maps and routes to help navigation.
Medical Product Sentiment Analysis from Physician Transcripts
Converted physician interview audio to text transcripts and then performed sentiment analysis using large language models (LLM) with 83% accuracy. Designed heuristics to go beyond sentiment analysis and tag physicians’ characteristics to predict their prescription practices and thereby predict actual product market share with Zero-Shot Accuracy of 71%.
Metadata Marketplace
Engineered a robust ETL pipeline leveraging Airflow to generate, enrich, and publish metadata from structured data sources to a knowledge graph, facilitating critical insights for over 1000 employees daily. Implemented data cataloging techniques to organize and classify metadata, enhancing accessibility for downstream applications. Collaborated with stakeholders to define 250 metadata standards and ontologies. Designed and implemented monitoring and alerting mechanisms within Kibana to track metadata quality and performance metrics, enabling proactive maintenance and troubleshooting with 96% uptime.ted an IoT application solution to measure real-time noise levels in the city and flag down noisy zones(exceeeding 55db) to help raise awareness and reduce noise pollution in urban areas. Utilized the Eclipse Paho library to provide MQTT client implementation for communication with the IoT app.
Snowflake ETL Pipeline
Developed end-to-end ETL pipeline leveraging Snowflake’s cloud-native data warehouse capabilities with Azure. Implemented automated data processing workflows using Apache Airflow, ensuring robust scheduling and monitoring. Leveraged PySpark for complex data transformations and processing tasks, optimizing performance by 39%.
Decibel- Noise Pollution Detection
Formulated an IoT application solution to measure real-time noise levels in the city and flag down noisy zones(exceeeding 55db) to help raise awareness and reduce noise pollution in urban areas. Utilized the Eclipse Paho library to provide MQTT client implementation for communication with the IoT app.
Some Awards and Achievements
Hackathon Wins
*
Created an application that screens resumes and takes preliminary HR interviews via a chat bot for applicant screening. Recruiters can also connect to the pool of suitable candidates via mail, chat or via our video calling facility.
@ Recruit-a-thon 2020- Recruitment Solutions Hackathon
Winner among 200+ teams
*
The project I developed was an IOT solution to solve the problems of potholes and noise pollution in urban regions of Mumbai.
@ Mumbai Hackathon 2019- an open-source hackathon
Winner Across the participants in Mumbai
*
Awarded first place in the maiden installment of IBM Hack_Challenge, in 2019 for making a AI powered Friend Recommendation system using personality analysis.
@ IBM Hack Challenge 2019
Winner in the Friend Affinity Finder Problem Statement
