Skip to content
View nehashastri's full-sized avatar

Highlights

  • Pro

Block or report nehashastri

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
nehashastri/README.md

👋 Hi there, I'm Neha Shastri!

Welcome to my GitHub! I'm Neha—a data enthusiast and applied AI/ML engineer who loves building things - LLM applications, RAG-style retrieval systems, automation workflows, and end-to-end deployable ML pipelines. With a background in Computer Science & Engineering and hands-on experience in data science, machine learning, and cloud technologies, I’m passionate about designing workflows, scaling data pipelines in Python, and turning messy data into intelligent, user-facing systems that solve real problems.

Currently, I’m pursuing my Master’s in Business Analytics at BU Questrom, specializing in Data & Methods, where I focus on applied machine learning, deep learning, NLP, and experimentation. I’m fortunate to be supported by the Director’s Achievement Scholarship and am actively engaged in case competitions and collaborative analytics projects. Most of my work sits at the intersection of Python, LLMs, Airflow, GCP, and scalable ML pipelines, always centered around creating reliable, deployable systems

What I’m working on now:

  • LLM-powered RAG pipelines (summarization, semantic search, retrieval, feature extraction)
  • Airflow + GCP + BigQuery ETL pipelines for credit-risk intelligence (daily records, idempotent DAGs, automated validation)

I love roles where I can experiment, automate, and build cool things - whether it’s running controlled experiments, designing ML-powered features, or creating intelligent workflows with LLMs. If you're working in the space of experimentation, applied ML, AI systems, or data-driven automation, I’d be happy to connect.


Technical Skills

  • 🌟 Programming: Python (Pandas, Scikit-learn, TensorFlow, PyTorch, Statsmodels, HuggingFace), SQL, Git
  • ML & Product Analytics: A/B Testing, Causal Inference, Experimentation Design, Customer Segmentation, Churn Prediction, Recommender Systems
  • LLM & GenAI: Transformers, BERT, RoBERTa, Prompt Engineering, Fine-tuning, Retrieval-Augmented Generation (RAG), LangChain
  • Big Data & Cloud: Hadoop, Apache Spark, AWS, GCP, Google BigQuery
  • Production & Monitoring: MLflow, Model A/B Testing, Feature Stores, Data Validation, Pipeline Orchestration (Airflow, Prefect)
  • Tools: Tableau, Power BI, Git, Jupyter Notebooks, Excel

Outside the world of data:

  • I’m a trained singer and was part of a rock band
  • I have a huge sweet tooth (chocolate and ice cream lover!)
  • Love exploring new places and cultures

How to Reach Me

📧 Email: nehags@bu.edu
💼 LinkedIn: linkedin.com/in/nehagshastri/


Thanks for stopping by! Feel free to explore my repositories or drop a message if you'd like to collaborate!

Pinned Loading

  1. Time-Series-Forecasting-A-15-Day-Practical-Study-Plan Time-Series-Forecasting-A-15-Day-Practical-Study-Plan Public

    I was unable to find a comprehensive resource that covers time series forecasting from the fundamentals through to model development and deployment. This repository aims to consolidate everything I…

    Jupyter Notebook 6

  2. Computer-Vision-Determining-Where-Eyes-Look Computer-Vision-Determining-Where-Eyes-Look Public

    A computer vision project that detects whether a person is looking directly at the camera or not, using deep learning and real-time video input.

    Python 1

  3. MinneMUDAC-2025-Mentorship-Matching MinneMUDAC-2025-Mentorship-Matching Public

    This repository contains our solution to the 2025 **MinneMUDAC Data Science Challenge**, in partnership with **Big Brothers Big Sisters (BBBS)**. Our goal: uncover what makes a mentorship match suc…

    Jupyter Notebook 1

  4. Retail-Customer-Behavior-Analysis Retail-Customer-Behavior-Analysis Public

    The targeted problem this project intends to solve is increasing sales on e-commerce platforms. Our objectives include identifying unusual patterns that may indicate fraud, grouping customers based…

    Jupyter Notebook 1 1

  5. Amazon-Reviews-Consumer-Sentiment-and-Trends Amazon-Reviews-Consumer-Sentiment-and-Trends Public

    This project analyzes 2023 Amazon reviews in the Clothing, Shoes & Jewelry category to uncover emerging trends and key factors influencing review ratings.

    Jupyter Notebook 1

  6. Analysis-of-IMDB-dataset---TV-shows-Movies-and-more Analysis-of-IMDB-dataset---TV-shows-Movies-and-more Public

    This project explores the relationship between film ratings, genres, and box office performance from 2006 to 2020.

    Jupyter Notebook 1