Skip to content
View DhritiAM's full-sized avatar
1 follower · 0 following

Achievements

Achievement: Arctic Code Vault ContributorAchievement: Pull Shark

Achievements

Achievement: Arctic Code Vault ContributorAchievement: Pull Shark

Block or report DhritiAM

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
DhritiAM/README.md

Hi there 👋

📌 Project 1: Toastmasters RAG System

Problem:
Toastmasters members often struggle to find accurate, up-to-date information scattered across multiple documents such as pathways, roles, and club processes.

What I built:
A Retrieval-Augmented Generation (RAG) system that provides document-grounded answers, improving accuracy and reducing hallucinations in responses.

Key features:

  • Three-stage pipeline: Ingestion → Retrieval & Generation → Evaluation
  • Query classification for improved retrieval routing
  • Metadata-driven filtering to narrow search scope
  • Reranking to improve relevance of retrieved documents

Tech: Python, LLMs, RAG, Vector Search
🔗 GitHub

📌 Project 2: ML Foundations

Goal:
Develop strong intuition for data behavior, predictive modeling, and generalization through hands-on implementation of core ML techniques and exploratory analysis.

What I worked on:

  • Exploratory Data Analysis (NYC Taxi Dataset):
    Analyzed large-scale trip data to uncover financial and temporal patterns.

    • Data preparation: sampling, cleaning, preprocessing
    • Analysis: revenue trends, peak demand hours, seasonal effects
    • Insights: operational and business strategies to optimize profitability

  • Linear Regression (Car Price Prediction):
    Built predictive models for car prices and studied the effect of regularisation.

    • Applied Ridge and Lasso to handle multicollinearity
    • Used regularisation for feature selection and improved generalization

  • Logistic Regression (Employee Attrition Prediction):
    Modeled employee attrition as a classification problem.

    • Data preprocessing and feature selection
    • Model training and evaluation
    • Intuitive data visualisations to interpret predictions and decision boundaries

Focus areas:

  • Bias–variance trade-off and overfitting
  • Role of regularisation in controlling model complexity
  • Translating data patterns into interpretable insights
  • Using data visualisation as a reasoning and diagnostic tool

Tech: Python, NumPy, Pandas, Matplotlib, Scikit-learn

🔗 Repos:

Popular repositories Loading

  1. EDA-NYC-taxi EDA-NYC-taxi Public

    This project presents an exploratory data analysis (EDA) of the NYC Taxi dataset (2023).

    Jupyter Notebook 1

  2. build_llm_from_scratch build_llm_from_scratch Public

    Building the fundamental blocks of LLMs from scratch

    Jupyter Notebook 1

  3. car_price_prediction_linear_regression car_price_prediction_linear_regression Public

    This project builds machine learning models to predict car prices. We experiment with Linear Regression, and also apply Ridge and Lasso regularization to handle multicollinearity and feature select…

    Jupyter Notebook 1

  4. logistic_regression_employee_attrition logistic_regression_employee_attrition Public

    Predict employee attrition using logistic regression, explore feature importance, and evaluate model performance with confusion matrix, sensitivity, specificity and ROC analysis.

    Jupyter Notebook 1

  5. RAG_toastmasters RAG_toastmasters Public

    A RAG-based tool designed to help Toastmasters members easily access accurate, document-grounded information about pathways, roles, and club processes

    Python 1

  6. PESU-IO-SUMMER PESU-IO-SUMMER Public

    Jupyter Notebook