Dhruv Miyani
Dhruv Miyani
Machine Learning Enginner |
Data Scientist
Education
Masters of Science In Artificial Intelligence @ Khoury College -NEU
Bachelor of Science In Information technology - 🥈 Secured Second Rank
Algorithms
Neural Networks
LLMs
MLOps
Object Oriented Design
AI-HCI
Linear Algebra
Web Development
Java Programming
Database System
Education
Masters of Science In Artificial Intelligence @ Khoury College -NEU
Bachelor of Science In Information technology - 🥈 Secured Second Rank
Algorithms
Neural Networks
LLMs
MLOps
Object Oriented Design
AI-HCI
Linear Algebra
Web Development
Java Programming
Database System
01
Algorithms
LLMs
03
Neural Networks
02
MLOps
04
Education
Masters of Science In Artificial Intelligence @ Khoury College -NEU
Bachelor of Science In Information technology - 🥈 Secured Second Rank
Algorithms
Neural Networks
LLMs
MLOps
Object Oriented Design
AI-HCI
Linear Algebra
Web Development
Java Programming
Database System
Work Experience
Machine Learning Engineer @ Monit
Optimized ML systems using AWS Ultra clusters for high-latency, developing NLP and Quantitative models for bank name & account classification, improving accuracy by 8.6%, enhancing performance and cost efficiency, and deploying on the insight engine for real-time predictions.
Enhanced model interpretability by leveraging feature importance analysis, and developed an interactive ExplainerDashboard to empower the product sales team and decision-makers with actionable insights.
Accuracy Improvement
Accuracy Improvement
8.6%
Machine Learning Engineer @ Monit
Optimized ML systems using AWS Ultra clusters for high-latency, developing NLP and Quantitative models for bank name & account classification, improving accuracy by 8.6%, enhancing performance and cost efficiency, and deploying on the insight engine for real-time predictions.
Enhanced model interpretability by leveraging feature importance analysis, and developed an interactive ExplainerDashboard to empower the product sales team and decision-makers with actionable insights.
Accuracy Improvement
Accuracy Improvement
8.6%
Machine Learning Engineer @ Monit
Optimized ML systems using AWS Ultra clusters for high-latency, developing NLP and Quantitative models for bank name & account classification, improving accuracy by 8.6%, enhancing performance and cost efficiency, and deploying on the insight engine for real-time predictions.
Enhanced model interpretability by leveraging feature importance analysis, and developed an interactive ExplainerDashboard to empower the product sales team and decision-makers with actionable insights.
Accuracy Improvement
Accuracy Improvement
8.6%
ML Reserach Assistant @
D'Amore-McKim School of Business -NEU
Implemented end-to-end machine learning model lifecycles, Applied regression and clustering algorithms to uncover insights from financial data, informing decisions on financial performance and its impact on forced labor.
Data pipelines using Google Cloud Platform, automating analysis of forced labor-risks in supply chains. Reducing manual work by 200+ hours.
Automation
200+ Hours
ML Reserach Assistant @
D'Amore-McKim School of Business -NEU
Implemented end-to-end machine learning model lifecycles, Applied regression and clustering algorithms to uncover insights from financial data, informing decisions on financial performance and its impact on forced labor.
Data pipelines using Google Cloud Platform, automating analysis of forced labor-risks in supply chains. Reducing manual work by 200+ hours.
Automation
200+ Hours
Junior Data Scientist @ Veloc -Surat
Extracted and refined customer review datasets through text preprocessing, including regex and normalization,
Built, tested, and optimized customer segmentation & recommendation models.
• Synthesized complex data insights and conducted A/B testing to validate findings, effectively communicating 6 critical customer pain points to stakeholders
Product Insights
6+
Junior Data Scientist @ Veloc -Surat
Extracted and refined customer review datasets through text preprocessing, including regex and normalization,
Built, tested, and optimized customer segmentation & recommendation models.
• Synthesized complex data insights and conducted A/B testing to validate findings, effectively communicating 6 critical customer pain points to stakeholders
Product Insights
6+
Teaching Assistant
Engaged 150+ undergraduates in 3 classes with an emphasis on Information Presentation and Visualization, honing their skills in technical communication and complex concept understanding.
• Hosted 13+ practical data visualization sessions leveraging JavaScript, D3.js, matplotlib, Vega-Altair, and Tableau,
equipping students with hands-on experience. Mentored students for data visualization projects, providing personalized technical support to foster a rapid learning environment and demonstrating strong interpersonal and troubleshooting skills.
13+
Session Hosted
Teaching Assistant
Engaged 150+ undergraduates in 3 classes with an emphasis on Information Presentation and Visualization, honing their skills in technical communication and complex concept understanding.
• Hosted 13+ practical data visualization sessions leveraging JavaScript, D3.js, matplotlib, Vega-Altair, and Tableau,
equipping students with hands-on experience. Mentored students for data visualization projects, providing personalized technical support to foster a rapid learning environment and demonstrating strong interpersonal and troubleshooting skills.
13+
Session Hosted
Skills
Python
C
SQL
JavaScript
PyTorch
Pandas
PEFT
RAG
LangChain
Amazon Redshift Serverless
Amazon Redshift Serverless
Amazon Redshift Serverless
SageMaker
Airflow
Docker
Explainer Dashboard
+ More
+ More
+ More
Google Cloud Platform
Deployment Automation
Deployment Automation
Deployment Automation
Projects
PedalOps
This project aims to predict the demand for BlueBikes using historical data and station information. The goal is to deploy an MLOps pipeline that automates data ingestion, model training, deployment, and retraining.
RAG Based Clinical Language Model
• Conducted a performance study by Implementing Retrieval-Augmented Generation (RAG) with GPT-3.5 and a fine-tuned ClinicalBERT model on Nosocomial Risk Datasets.
• Demonstrated through research the subtle yet significant edge of RAG-integrated LLMs over fine-tuned specialized models in clinical domain-specific tasks, Utilized LlamaIndex to query clinical records, enhancing the efficiency and accuracy of data retrieval for model training, and evaluated the results using RAGAS score.
Hate Speech Detection In Low Resource Language
• Designed and implemented a six-class hate speech detection system for Gujarati-language social media text.
• Collected and preprocessed data from social media platforms, leveraging Regex techniques for text cleaning, advanced tokenization, and TF-IDF vectorization for feature extraction.
• Performed experiments utilizing multiple machine learning algorithms including Logistic Regression, K-Nearest Neighbors (KNN), Naive Bayes, and SGD.
THANKS FOR STOPPING BY!