Ruchi Bagwe
I am a performance-driven Software Engineer with a strong background in Machine Learning and Data Science Technologies. I have eight years of data-driven industrial experience. My research comes broadly under Natural Language Processing and relates to Natural Language Generation, Machine translation, Text Analysis, and cognitive science. Currently, I am working at Qualcomm as a Senior Machine Learning Engineer.I am working on optimizing the Deep Learning Algorithms to make Qualcomm Snapdragon chips more power-efficient.I seek not only to learn new ways to solve problems, but also to learn from other experts on the team. I believe the combination of these strengths can help produce tangible outcomes in the fields of data science and machine learning.
Skills
Languages, Operating Systems & Tools
- Python
- javascript
- git
- NodeJS
- PySpark
- linux
- Java
Machine Learning
- Natural Language Processing
- Image Generation
- Text analysis
- Speech To Text
- Data Mining Forecasting
- Time Series Analysis
- Tensorflow
- PyTorch
- Scikit
- NLTK
- Keras
- Genism
Containers & Cloud
- Google Cloud Platform
- Apache Spark
Data Management
- MongoDB
- SAP HANA
- Cassandra
- Microsoft SQL Server
- Oracle
- Postgres
- NoSQL
Basic Knowledge of :
- AWS
- Airflow
- Kafka
- Scala
- MapReduce
Product description classification using NER
In this project, the product is classified by performing entity recognition from product descriptio using NER. The laptops data was scrapped from the eBay website using Beautifulsoup. Entities are the laptop description (Product Brand, Product Model, Hard disk type/size, display type, etc.) available on the eBay site.
Wait! There's more..
See all Projects for more examples!
IEEE Paper - Natural Language Controlled Real-Time Object Recognition Framework for Household Robot
2021 IEEE 11th Annual Computing and Communication Workshop and Conference (CCWC)
March 2021
![]()
IEEE Paper - Cortically Coupled Generative Adversarial Network for Target Image Retrieval in Rapid Image Search
2020 IEEE Second International Conference on Cognitive Machine Intelligence (CogMI)
January 2021
![]()
IEEE Paper - Automatic Numerical Question Answering on Table Using BERT-GNN
2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON)
December 2020
![]()
IEEE Paper - Automated Radar Signal Analysis Based on Deep Learning
2020 10th Annual Computing and Communication Workshop and Conference (CCWC)
January 2019
![]()
Experience
Senior Machine Learning Engineer
Qualcomm
- Worked on optimizing the computer vision models for object detection, segmentation, and classification for the mobile application to achieve higher inference time and optimize power consumption
- Working on data preparation and augmentation for Automatic Speech Recognition application, which involves deep-learning-based(NLP and Computer Vision) speaker feature extraction, speaker verification, speaker identification, and speaker diarisation. .
May 2021 - Present
Machine Learning Intern
ScriptChain LLC
- Developed a plugin platform in NodeJS and JavaScript to aggregate and transform the clinical data from different healthcare provider networks to provide insight into patient healthcare hosted on the web portal. Built operational data pipeline to process and analyses healthcare data from Data Lake to the Google Cloud Platform
- Fine-tuned BERT to build an NLP text search for clinical notes, and used other libraries like MetaMap, PyTorch, Gensim
- Worked on building a health risk calculator using Random Forest, Naïve Bayes, and CNN on the healthcare data
May 2020 - July 2020
Graduate Research Assistant
California State University, Fullerton
- Worked on robot learning for intuitive human-robot interaction using Natural Language programming and Computer Vision. Developed a Natural Language controlled real-time YOLOv4 based object recognition and classification framework for household robot for vetrans funded project (http://news.fullerton.edu/2020/11/technology-researchers-develop-robotic-arm-for-blind-veterans/)
- Extracted features from EEG data and mapped it into Gaussian distribution, built encoder and decoder using deep learning algorithms CNN, LSTM, and GAN for object generation
- Developed the the automatic numerical question answering system on Tables using BERT and Graph Neural Network. Till now, achieved 23% accuracy on solving numerical questions ..
July 2019 - Dec 2020
Data Scientist
Tata Consultancy Services, Brussels, Belgium
- Designed and developed Spark ETL applications using Spark Dataframes, SQL, and RDDs to extract, manipulate and aggregate the transactional data from customer-facing database to HDFS.
- Worked on data gathering, preprocess, and analysis for developing an AI-Enabled Customer Service Channel project, which focused on improving customer services by analyzing customer issues and satisfaction data using NLP
- Analyzed and preprocessed unstructured customer services tickets dataset using PySpark, NLTK, and Python, developed a custom data model for automatic ticket categorization.
- Built end-to-end machine learning data pipeline to translate DS/ML metrics to business KPIs
- Showcased leadership qualities by engaging in requirements gathering and project scope discussions with customers and different functional teams. Coordinated with the offshore technical team and guided them through the development and testing phases
August 2016 – December 2018
SOFTWARE ENGINEER
Tata Consultancy Services, Mumbai, India
- Performed role of software developer and data engineer for TCS in house finance payroll project. The role included creating the data pipelines from application databases to Data Warehouse and developing data models, stored procedures, and functions
- Improved DataStage effectiveness by tuning the data workflow, identified automation opportunities, re-engineered data pipeline processes, which reduced the load/run by 7 hours
- Worked on performance tuning, analysis, and response time reduction techniques in SQL. Optimized SQL code and developed SQL stored procedures on incremental data, Achieving a 23% improvement in application response time. Automated database health-check operation using Python, Unix Scripting, PLSQL, and HTML.
December 2010 – August 2016
Education
Jan-2019 - Jan-2021
June-2006 - June-2010