Darshan Kholakiya
Mumbai
Summary
Applied Data Scientist specializing in Machine Learning and Generative AI. Experienced in building and evaluating LLM-powered applications, designing classification models, and analyzing user interactions to improve AI decision accuracy and system performance.
Overview
4
4
years of professional experience
1
1
Certification
Work History
Data Scientist
Global Nodes
Gurugram
09.2024 - Current
- Architected and deployed an AI Assistant for post-surgery patients with 48 specialized LLM Chains using Python, LangChain, and AWS services (DynamoDB, CloudWatch): live in production.
- Analyzed patient conversations, engineered features and trained a Random Forest classifier achieving 87% accuracy to automate classification of conversation in an LLM healthcare assistant.
- Performed performance analysis on LLM response latency, identified initialization bottlenecks and redesigned inference workflow (singleton LLM loading and prompt restructuring), cutting average response time from 26s to 14s (~46%) and significantly improving conversational usability.
- Injected medical text into Postgres Vector Database for Retrieval-Augmented Generation (RAG), improving patient recommendation accuracy.
- Developed a Streamlit app for SMEs to test the AI Chat across multiple LLM models and surgery types.
- Collaborated with cross-functional teams to integrate AI solutions into production, facilitating seamless deployment of generative AI applications.
- Designed an FDA-submission-ready Model Migration Plan to ensure zero behavioral regression on model updates.
- Optimized migration verification via multithreading, saving ~8.36 hours of runtime (~74% reduction).
Junior Data Scientist
Global Nodes
Mumbai
02.2024 - 09.2024
- Designed and implemented AI-powered chat functionality for a healthcare product, leveraging LLMs and advanced prompt engineering techniques.
- Created Power BI dashboards for monitoring of AI chat performance, enabling issue detection and resolution.
- Conducted thorough unit testing of designed components, ensuring robustness and reliability of high-quality code.
- Collaborated with team members and stakeholders, facilitating effective communication to achieve impactful project results.
- Contributed ideas and insights during problem-solving and brainstorming sessions, enhancing project outcomes and product capabilities.
- Continuously expanded knowledge and skills in AI and data science through self-learning and staying up-to-date with the latest industry trends and best practices.
Data Science Intern
BugendaiTech
Mumbai
06.2023 - 12.2023
- Conceptualized and designed an innovative system to automate and personalize project documentation or summarization of meetings notes, revolutionizing the project management process, which increased team efficiency by 50%.
- Prioritized data privacy as a core objective of the project while concurrently achieving increase in team productivity and a remarkable reduction of making project documentation time by 60%.
- Developed and lead the implementation of the 'Project Documentation Generator' using Large Language Models.
- Precisely fine-tuned a pre-trained model, ensuring data quality aligned with the organization's specific requirements.
- Pioneered the design of the User Interface on Gradio, enabling real-time streaming of the output.
Data Analyst Intern
DoWell Research India
Mumbai
08.2022 - 02.2023
- Leveraged Google Sheets and Looker Studio to transform raw global indicator data into 20+ insightful reports, including interactive visualizations like filled maps and funnel charts, for stakeholders across departments.
- Collaborated with data analysts to ensure data integrity through quality control measures, enabling accurate analysis that drove strategic business decisions and informed multiple teams' reporting.
- Analyzed data sets to identify trends and insights for market research projects.
- Collaborated with team members to develop comprehensive reports and presentations.
- Utilized statistical tools to perform data modeling and forecasting activities.
Education
BSc - Data Science
KES Shroff College
Mumbai, India05.2023
H.S.C. - Science
Shri T.P. Bhatia College
Mumbai, India03.2020
S.S.C. -
Rustomjee International School
Mumbai, India04.2018
Skills
- Python
- SQL
- AWS
- NLP
- LLM
- Generative AI
- Machine Learning
- Deep Learning
- A&B Testing
- Prompt Engineering
- Prompt Performance Analysis
- Experiment Tracking
- Reusable Notebooks
- Data Cleaning
- Data Preprocessing
- Exploratory Data Analysis
- Vector Database
- RAG
- LangGraph
- Langchain
- Gradio
- Streamlit
- Streamlit Cloud
- Looker Studio
- Tableau
- Power BI
- GitHub
- Git
Certification
- Data Scientist (TryCatch Classes)
- Introduction to Generative AI (Google)
- Data Visualisation: Empowering Business with Effective Insights (Tata)
- Data Analytics and Visualization Virtual Experience (Accenture)
- Data Analytics Consulting Virtual Internship (KPMG)
- Python Basics (HackerRank)
- SQL Basics (HackerRank)
- Data Analysis with Python (IBM Developer Skills Network)
- Data Science 101 (IBM Developer Skills Network)
Projects
- WhatsApp Chat Analyzer, Data Analysis & Preprocessing, Data Visualization, Sentiment Analysis, This tool will help users understand their WhatsApp chat’s patterns, frequency of messages exchanged, and more.
- Laptop Price Predictor, Machine Learning, Web Scraping, HTML, Streamlit, Designed to predict the prices of laptops and provide 3 recommendations based on various features and specifications.
- Asian Landmark Detection, TensorFlow Hub, Image Processing, Geo Spatial Visualization, The application takes an image as an input and predicts the name of the Asian Landmark with the geolocation address.
- Auto Tech News, Web Scraping, HTML, CSS, Flask, This application web scraps the latest technology news and presents the top 6 news articles in a visually designed template, making it easy for users to access and download the latest news.
- PM Modi Speech Comparison (2020 vs 2021), NLP, Python, Data Presentation, Jupyter Notebook, Analyze and gain insights from PM Modi’s Speech of 15th August 2020 vs 2021. Utilize NLP techniques such as Stop Word Removal, Lemmatization, Sentiment Analysis, Name Entity Recognition, Topic Modeling.
Timeline
Data Scientist
Global Nodes
09.2024 - Current
Junior Data Scientist
Global Nodes
02.2024 - 09.2024
Data Science Intern
BugendaiTech
06.2023 - 12.2023
Data Analyst Intern
DoWell Research India
08.2022 - 02.2023
BSc - Data Science
KES Shroff College
H.S.C. - Science
Shri T.P. Bhatia College
S.S.C. -
Rustomjee International School